From csnet_gateway Tue Sep 16 06:45:13 1986 Date: Tue, 16 Sep 86 06:45:08 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #180 Status: R AIList Digest Tuesday, 16 Sep 1986 Volume 4 : Issue 180 Today's Topics: Administrivia - Resumption of Service, AI Tools - C, Expert Systems - Matching, Philosophy - Argumentation Style & Sports Analogy, Physiology - Rate of Tissue Replacement ---------------------------------------------------------------------- Date: Tue 16 Sep 86 01:28:28-PDT From: Ken Laws Reply-to: AIList-Request@SRI-AI.ARPA Subject: Resumption of Service I'm back from vacation and almost have the mail streams under control again. This issue clears out "old business" messages relating to the discussions in early August. I'll follow with digests flushing the accumulated queries, Usenet replies, news items, conference and seminar abstracts, and bibliographic citations -- spread out a bit so that I'm not deluged with mailer bounce messages from readers who have dropped without notification. Incidentally, about 30 people signed up for direct distribution this month despite the inactivity of the list. (Most of the additions for the last year have been on BITNET, often in clusters as new universities join the net or become aware of the Arpanet digests. Most Arpanet and CSNet sites are now using bboards and redistribution lists or are making use of the Usenet mod.ai/net.ai distribution.) I plan to pass along only an abbreviated announcement for conferences that have already been announced in the NL-KR, IRList, or Prolog lists -- you can contact the message author if you need the full text. (Note that this may reduce the yield of keyword searches through the AIList archive; future historians will have to search the other lists to get a full picture of AI activity. Anyone building an intelligent mail-screening system should also incorporate cross-list linkages. Any such screening system that can understand and coordinate these message streams deserves a Turing award.) -- Ken Laws ------------------------------ Date: Wed, 20 Aug 86 10:07:49 edt From: cdx39!jc%rclex.UUCP@harvard.HARVARD.EDU Subject: Re: Reimplementing in C > I've been hearing and seeing something for the past couple years, > something that seems to be becoming a folk theorem. The theorem goes > like this: > Many expert systems are being reimplemented in C. > I'm curious what the facts are. [I program in C, and have reached the conclusion that most AI programming could be done in that language as easily as in LISP if libraries of list-oriented subroutines were available. (They needn't be consed lists -- I use dynamically allocated arrays.) You do have to worry about storage deallocation, but that buys you considerable run-time efficiency. You also lose the powerful LISP debugging environment, so fill your code with lots of argument checks and ASSERTs. Tail recursion isn't optimized, so C code should use iteration rather than recursion for most array-based list traversals. Data-driven and object-oriented coding are easy enough, but you can't easily build run-time "active objects" (i.e., procedures to be applied to message arguments); compiled subroutines have to do the work, and dynamic linking is not generally worth the effort. I haven't tried much parsing or hierarchy traversal, but programs such as LEX, YACC, and MAKE show that it can be done. -- KIL] Well, now, I don't know about re-implementing in C, but I myself have been doing a fair amount of what might be called "expert systems" programming in C, and pretty much out of necessity. This is because I've been working in the up-and-coming world of networks and "intelligent" communication devices. These show much promise for the future; unfortunately they also add a very "interesting" aspect to the job of an application (much less a system) programmer. The basic problem is that such comm devices act like black boxes with a very large number of internal states; the states aren't completely documented; those that are documented are invariably misunderstood by anyone but the people who built the boxes; and worst of all, there is usually no reliable way to get the box into a known initial state. As a result, there is usually no way to write a simple, straightforward routine to deal with such gadgets. Rather, you are forced to write code that tries to determine 1) what states a given box can have; 2) what state it appears to be in now; and 3) what sort of command will get it from state X to state Y. The debugging process involves noting unusual responses of the box to a command, discussing the "new" behavior with the experts (the designers if they are available, or others with experience with the box), and adding new cases to your code to handle the behavior when it shows up again. One of the simplest examples is an "intelligent ACU", which we used to call a "dial-out modem". These now contain their own processor, plus sufficiently much ROM and RAM to amount to small computer systems of their own. Where such boxes used to have little more than a status line to indicate the state of a line (connected/disconnected), they now have an impressive repertoire of commands, with a truly astonishing list of responses, most of which you hope never to see. But your code will indeed see them. When your code first talks to the ACU, the responses may include any of: 1. Nothing at all. 2. Echo of the prompt. 3. Command prompt (different for each ACU). 4. Diagnostic (any of a large set). Or the ACU may have been in a "connected" state, in which case your message will be transmitted down the line, to be interpreted by whatever the ACU was connected to by the most recent user. (This recursive case is really fun!:-) The last point is crucial: In many cases, you don't know who is responding to your message. You are dealing with chains of boxes, each of which may respond to your message and/or pass it on to the next box. Each box has a different behaviour repertoire, and even worse, each has a different syntax. Furthermore, at any time, for whatever reason (such as power glitches or commands from other sources), any box may reset its internal state to any other state. You can be talking to the 3rd box in a chain, and suddenly the 2nd breaks in and responds to a message not intended for it. The best way of handling such complexity is via an explicit state table that says what was last sent down the line, what the response was, what sort of box we seem to be talking to, and what its internal state seems to be. The code to use such info to elicit a desired behavior rapidly develops into a real piece of "expert-systems" code. So far, there's no real need for C; this is all well within the powers of Lisp or Smalltalk or Prolog. So why C? Well, when you're writing comm code, you have one extra goodie. It's very important that you have precise control over every bit of every character. The higher-level languages always seen to want to "help" by tokenizing the input and putting the output into some sort of standard format. This is unacceptable. For instance, the messages transmitted often don't have any well-defined terminators. Or, rather, each box has its own terminator(s), but you don't know beforehand which box will respond to a given message. They often require nulls. It's often very important whether you use CR or LF (or both, in a particular order). And you have to timeout various inputs, else your code just hangs forever. Such things are very awkward, if not impossible to express in the typical AI languages. This isn't to say that C is the world's best AI language; quite the contrary. I'd love to get a chance to work on a better one. (Hint, hint....) But given the languages available, it seems to be the best of a bad lot, so I use it. If you think doing it in C is weird, just wait 'til you see it in Ada.... ------------------------------ Date: 2 Sep 86 08:31:00 EST From: "CLSTR1::BECK" Reply-to: "CLSTR1::BECK" Subject: matching Mr. Rosa, is correct in saying that "the obstacles to implementation are not technological," since this procedure is currently being implemented. See "matches hit civil servants hardest" in the august 15, 1986 GOVERNMENT COMPUTER NEWS. "Computer Matching/Matches" is defined as searching the available data for addresses, financial information, specific personal identifiers and various irregularities". The congressional Office of Technology Assessment has recently issued a report, "Electronic Record Systems and Individual Privacy" that discusses matching. My concern with this is how will the conflicting rules of society be reconciled to treat the indiviual fairly. Maybe the cash society and anonymous logins will become prevalent. Do you think that the falling cost of data will force data keepers to do more searches to justify their existence? Has there been any discussion of this topic? peter beck ------------------------------ Date: Tue, 12 Aug 86 13:20:20 EDT From: "Col. G. L. Sicherman" Subject: Re: philosophy articles > >Out of curiosity I hunted up [Jackson, "What Mary Didn't Know," _J. > >of Philosophy_ 83(1986) 291-295] on the way back from lunch. > >It's aggressive and condescending; any sympathy I might have felt for > >the author's argument was repulsed by his sophomoric writing. I hope it's > >not typical of the writing in philosophy journals. > > I don't quite understand what "aggressive and condescending" or > "sophomoric writing" have to do with philosophical argumentation. > One thing that philosophers try not to do is give ad hominem arguments. > A philosophical arguement stands or falls on its logical merits, not its > rhetoric. That's an automatic reaction, and I think it's unsound. Since we're not in net.philosophy, I'll be brief. Philosophers argue about logic, terminology, and their experience of reality. There isn't really much to argue about where logic is concerned: we all know the principles of formal logic, and we're all writing sincerely about reality, which has no contradictions in itself. What we're really interested in is the nature of our exist- ence; the logic of how we describe it doesn't matter. One reason that Jackson's article irritated me is that he uses formal logic, of the sort "Either A or B, but not A, therefore B." This kind of argument insults the reader's intelligence. Jackson ought to know that nobody is going to question the soundness of such logic, but that all his opponents will question his premises and his definitions. More- over, he appears to regard his premises and definitions as unassailable. I call that stupid philosophizing. Ad-hominem attacks may well help to discover the truth. When the man with jaundice announces that everything is fundamentally yellow, you must attack the man, not the logic. So long as he's got the disease, he's right! ------------------------------ Date: Tue, 12 Aug 86 12:53:07 EDT From: "Col. G. L. Sicherman" Subject: Re: talk to the medium (from Risks Digest) > Whether he was talking about the broadcast or the computer industry, he > got the analogy wrong. Of course--that's what makes the analogy "stick." > If the subject is broadcasting, the sports analogy to a "programmer" > is the guy that makes the play schedules. Not exactly. McLuhan's "programmer" is the man who selects the content of the medium, not what computer people call a programmer. > ... But still, in computing, > a programmer bears at least partial responsibility for the computer's > (mis)behaviour. I agree. McLuhan is writing about not responsibility but responsiveness. Last Saturday I visited an apartment where a group of men and kids were shouting at the TV set during a football game. It's a natural response, and it would have been effective if TV were an interactive medium. If you dislike this posting, will you complain to the moderator? To the people who programmed netnews? To the editor of the New York _Times?_ Of course not; you must talk to the medium, not to the programmer! ------------------------------ Date: Wed, 20 Aug 86 10:06:56 edt From: cdx39!jc%rclex.UUCP@harvard.HARVARD.EDU Subject: Re: "Proper" Study of Science, Conservation of Info [The following hasn't any obvious AI, but it's interesting enough to pass along. Commonsense reasoning at work. -- KIL] > The ability to quantify and measure ... has profound implications ... > > ... A decade from now it's likely that none of our bodies > will contain EVEN A SINGLE ATOM now in them. Even bones are fluid in > biological organisms; ... OK, let's do some BOTE (Back Of The Envelope) calculations. According to several bio and med texts I've read over the years, a good estimate of the half-life residency of an atom in the soft portions of a mammal's body is 1/2 year; in the bones it is around 2 years. The qualifications are quite obvious and irrelevant here; we are going for order-of-magnitude figures. For those not familiar with the term, "half-life residency" means the time to replace half the original atoms. This doesn't mean that you replace half your soft tissues in 6 months, and the other half in the next six months. What happens is exponential: in one year, 1/4 of the original are left; in 18 months, 1/8 are left, and so on. Ten years is about 5 half-lives for the bones, and 20 for the soft tissues. A human body masses about 50 Kg, give or take a factor of 2. The soft tissues are primarily water (75%) and COH2; we can treat it all as water for estimating the number of atoms. This is about (50Kg) * (1000 KG/g) / (16 g/mole) = 3000 moles, times 6*10^23 gives us about 2*10^26 atoms. The bones are a bit denser (with fewer atoms per gram); the rest is a bit less dense (with more atoms per gram), but it's about right. For order-of-magnitude estimates, we would have roughly 10^26 atoms in each kind of tissue. In 5 half-lives, we would divide this by 2^5 = 32 to get the number of original atoms, giving us about 7*10^25 atoms of the bones left. For the soft tissues, we divide by 2^20 = 4*10^6, giving us about 2 or 3 * 10^20 of the original atoms. Of course, although these are big numbers, they don't amount to much mass, especially for the soft tissues. But they are a lot more than a single atom, even if they are off by an order of magnitude.. Does anyone see any serious errors in these calculations? Remember that these are order-of magnitude estimates; quibbling with anything other than the first significant digit and the exponent is beside the point. The only likely source of error is in the half-life estimate, but the replacement would have to be much faster than a half-year to stand a chance of eliminating every atom in a year. In fact, with the exponential-decay at work here, it is easy to see that it would take about 80 half-lives (2*10^26 = 2^79) to replace the last atom with better than 50% probability. For 10 years, this would mean a half-life residency of about 6 weeks, which may be true for a mouse or a sparrow, but I've never seen any hint that human bodies might replace themselves nearly this fast. In fact, we can get a good upper bound on how fast our atoms could be replaced, as well as a good cross-check on the above rough calculations, by considering how much we eat. A normal human diet is roughly a single Kg of food a day. (The air breathed isn't relevant; very little of the oxygen ends up incorporated into tissues.) In 6 weeks, this would add up to about 50 Kg. So it would require using very nearly all the atoms in our food as replacement atoms to do the job required. This is clearly not feasible; it is almost exactly the upper bound, and the actual figure has to be lower. A factor of 4 lower would give us the above estimate for the soft tissues, which seems feasible. There's one more qualification, but it works in the other direction. The above calculations are based on the assumption that incoming atoms are all 'new'. For people in most urban settings, this is close enough to be treated as true. But consider someone whose sewage goes into a septic tank and whose garbage goes into a compost pile, and whose diet is based on produce of their garden, hen-house, etc. The diet of such people will contain many atoms that have been part of their bodies in previous cycles, especially the C and N atoms, but also many of the O and H atoms. Such people could retain a significantly larger fraction of original atoms after a decade. Please don't take this as a personal attack. I just couldn't resist the combination of the quoted lines, which seemed to be a clear invitation to do some numeric calculations. In fact, if someone has figures good to more places, I'd like to see them. ------------------------------ End of AIList Digest ******************** From csnet_gateway Fri Sep 19 19:08:39 1986 Date: Fri, 19 Sep 86 19:08:30 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #181 Status: R AIList Digest Wednesday, 17 Sep 1986 Volume 4 : Issue 181 Today's Topics: Conferences - ACM Office Information Systems & IEEE Symposium on Logic Programming '86 ---------------------------------------------------------------------- Date: Mon, 15 Sep 86 23:14:51 edt From: rba@petrus.bellcore.com (Robert B. Allen) Subject: Conference on Office Information Systems - Brown U. ACM CONFERENCE ON OFFICE INFORMATION SYSTEMS October 6-8, 1968, Providence, R.I. Conference Chair: Carl Hewitt, MIT Program Chair: Stan Zdonik, Brown University Keynote Speaker: J.C.R. Licklider, MIT Distinguished Lecturer: A. van Dam, Brown University COIS is a major research conference on the design and use of computing systems for professional and knowledge workers. At this meeting, sessions and panels emphasize AI and organizational models of offices as sites for distributed information processing. Other themes include user interfaces, graphics, group cooperation, and object-oriented systems. For more information, call the Conference Registrar at Brown U. (401-813-1839), or send electronic mail to mhf@brown.CSNET. ------------------------------ Date: Tue, 9 Sep 86 23:50:34 MDT From: keller@utah-cs.ARPA (Bob Keller) Subject: Conference - SLP '86 We have requested, and the IEEE has agreed, that Symposium registrations be accepted at the "early" fee for a couple of more days, so please act immediately if you wish to exploit this rate. [Sorry for the delay -- AIList doesn't always function in real time. -- KIL] Hotel Reservations: phone 801-531-1000, telex 389434 The (nearly) final schedule: SLP '86 Third IEEE Symposium on LOGIC PROGRAMMING September 21-25, 1986 Westin Hotel Utah Salt Lake City, Utah SUNDAY, September 21 19:00 - 22:00 Symposium and tutorial registration MONDAY, September 22 08:00 - 09:00 Symposium and tutorial registration 09:00 - 17:30 TUTORIALS (concurrent) Please see abstracts later. George Luger Introduction to AI Programming in Prolog University of New Mexico David Scott Warren Building Prolog Interpreters SUNY, Stony Brook John Conery Theory of Parallelism, with Applications to University of Oregon Logic Programming 12:00 - 17:30 Exhibit set up time 18:00 - 22:00 Symposium registration 20:00 - 22:00 Reception TUESDAY, September 23 08:00 - 12:30 Symposium registration 09:00 Exhibits open 09:00 - 09:30 Welcome and announcements 09:30 - 10:30 INVITED SPEAKER: W. W. Bledsoe, MCC Some Thoughts on Proof Discovery 11:00 - 12:30 SESSION 1: Applications (Chair: Harvey Abramson) The Logic of Tensed Statements in English - an Application of Logic Programming Peter Ohrstrom, University of Aalborg Nils Klarlund, University of Aarhus Incremental Flavor-Mixing of Meta-Interpreters for Expert System Construction Leon Sterling and Randall D. Beer Case Western Reserve University The Phoning Philosopher's Problem or Logic Programming for Telecommunications Applications J.L. Armstrong, N.A. Elshiewy, and R. Virding Ericsson Telecom 14:00 - 15:30 SESSION 2: Secondary Storage (Chair: Maurice Bruynooghe) EDUCE - A Marriage of Convenience: Prolog and a Relational DBMS Jorge Bocca, ECRC, Munich Paging Strategy for Prolog Based Dynamic Virtual Memory Mark Ross, Royal Melbourne Institute of Technology K. Ramamohanarao, University of Melbourne A Logical Treatment of Secondary Storage Anthony J. Kusalik, University of Saskatchewan Ian T. Foster, Imperial College, London 16:00 - 17:30 SESSION 3: Compilation (Chair: Richard O'Keefe) Compiling Control Maurice Bruynooghe, Danny De Schreye, Bruno Krekels Katholieke Universiteit Leuven Automatic Mode Inference for Prolog Programs Saumya K. Debray, David S. Warren SUNY at Stony Brook IDEAL: an Ideal DEductive Applicative Language Pier Giorgio Bosco, Elio Giovannetti C.S.E.L.T., Torino 17:30 - 19:30 Reception 20:30 - 22:30 Panel (Wm. Kornfeld, moderator) Logic Programming for Systems Programming Panelists: Steve Taylor, Weizmann Institute Steve Gregory, Imperial College Bill Wadge A researcher from ICOT (sorry this is incomplete) WEDNESDAY, September 24 09:00 - 10:00 INVITED SPEAKER: Sten Ake Tarnlund, Uppsala University Logic Programming - A Logical View 10:30 - 12:00 SESSION 4: Theory (Chair: Jean-Louis Lassez) A Theory of Modules for Logic Programming Dale Miller University of Pennsylvania Building-In Classical Equality into Prolog P. Hoddinott, E.W. Elcock The University of Western Ontario Negation as Failure Using Tight Derivations for General Logic Programs Allen Van Gelder Stanford University 13:30 - 15:00 SESSION 5: Control (Chair: Jacques Cohen) Characterisation of Terminating Logic Programs Thomas Vasak, The University of New South Wales John Potter, New South Wales Institute of Technology An Execution Model for Committed-Choice Non-Deterministic Languages Jim Crammond Heriot-Watt University Timestamped Term Representation in Implementing Prolog Heikki Mannila, Esko Ukkonen University of Helsinki 15:30 - 22:00 Excursion THURSDAY, September 25 09:00 - 10:30 SESSION 6: Unification (Chair: Uday Reddy) Refutation Methods for Horn Clauses with Equality Based on E-Unification Jean H. Gallier and Stan Raatz University of Pennsylvania An Algorithm for Unification in Equational Theories Alberto Martelli, Gianfranco Rossi Universita' di Torino An Implementation of Narrowing: the RITE Way Alan Josephson and Nachum Dershowitz University of Illinois at Urbana-Champaign 11:00 - 12:30 SESSION 7: Parallelism (Chair: Jim Crammond) Selecting the Backtrack Literal in the AND Process of the AND/OR Process Model Nam S. Woo and Kwang-Moo Choe AT & T Bell Laboratories Distributed Semi-Intelligent Backtracking for a Stack-based AND-parallel Prolog Peter Borgwardt, Tektronix Labs Doris Rea, University of Minnesota The Sync Model for Parallel Execution of Logic Programming Pey-yun Peggy Li and Alain J. Martin California Institute of Technology 14:00 - 15:30 SESSION 8: Performance Redundancy in Function-Free Recursive Rules Jeff Naughton Stanford University Performance Evaluation of a Storage Model for OR-Parallel Execution Andrzej Ciepelewski and Bogumil Hausman Swedish Institute of Computer Science (SICS) MALI: A Memory with a Real-Time Garbage Collector for Implementing Logic Programming Languages Yves Bekkers, Bernard Canet, Olivier Ridoux, Lucien Ungaro IRISA/INRIA Rennes 16:00 - 17:30 SESSION 9: Warren Abstract Machine (Chair: Manuel Hermenegildo) A High Performance LOW RISC Machine for Logic Programming J.W. Mills Arizona State University Register Allocation in a Prolog Machine Saumya K. Debray SUNY at Stony Brook Garbage Cut for Garbage Collection of Iterative Programs Jonas Barklund and Hakan Millroth Uppsala University EXHIBITS: An exhibit area including displays by publishers, equipment manufacturers, and software houses will accompany the Symposium. The list of exhibitors includes: Arity, Addison-Wesley, Elsevier, Expert Systems, Logicware, Overbeek Enterprises, Prolog Systems, and Quintus. For more information, please contact: Dr. Ross A. Overbeek Mathematics and Computer Science Division Argonne National Laboratory 9700 South Cass Ave. Argonne, IL 60439 312/972-7856 ACCOMODATIONS: The Westin Hotel Utah is a gracious turn of the century hotel with Mobil 4-Star and AAA 5-Star ratings. The Temple Square Hotel, located one city block away, offers basic comforts for budget-conscious attendees. MEALS AND SOCIAL EVENTS: Symposium registrants (excluding students and retired members) will receive tickets for lunches on September 23, 24, and 25, receptions on September 22 and 23, and an excursion the afternoon of September 24. The excursion will comprise a steam train trip through scenic Provo Canyon, and a barbeque at Deer Valley Resort, Park City, Utah. Tutorial registrants will receive lunch tickets for September 22. TRAVEL: The Official Carrier for SLP '86 is United Airlines, and the Official Travel Agent is Morris Travel (361 West Lawndale Drive, Salt Lake City, Utah 84115, phone 1-800-621-3535). Special airfares are available to SLP '86 attendees. Contact Morris Travel for details. A courtesy limousine is available from Salt Lake International Airport to both symposium hotels, running every half hour from 6:30 to 23:00. The taxi fare is approximately $10. CLIMATE: Salt Lake City generally has warm weather in September, although evenings may be cool. A warm jacket should be brought for the excursion. Some rain is normal this time of year. SLP '86 Symposium and Tutorial Registration Coupon: Advance symposium and tutorial registration is available until September 1, 1986. No refunds will be made after that date. Send a check or money order (no currency will be accepted) payable to "Third IEEE Symposium on Logic Programming" to: Third IEEE Symposium on Logic Programming IEEE Computer Society 1730 Massachusetts Avenue, N.W. Washington, D.C. 20036-1903 [...] Symposium Registration: Advance On-Site IEEE Computer Society members $185 $215 Non-members $230 $270 Full-time student members $ 50 $ 50 Full-time student non-members $ 65 $ 65 Retired members $ 50 $ 50 Tutorial Registration: ("Luger", "Warren", or "Ostlund") Advance On-Site IEEE Computer Society members $140 $170 Non-members $175 $215 SLP '86 Hotel Reservation: Mail or Call: phone 801-531-1000, telex 389434 Westin Hotel Utah Main and South Temple Streets Salt Lake City, UT 84111 A deposit of one night's room or credit card guarantee is required for arrivals after 6pm. Room Rates: Westin Hotel Utah Temple Square Hotel single room $60 $30 double room $70 $36 Reservations must be made mentioning SLP '86 by August 31, 1986 to guarantee these special rates. SLP '86 TUTORIAL ABSTRACTS IMPLEMENTATION OF PROLOG INTERPRETERS AND COMPILERS DAVID SCOTT WARREN SUNY AT STONY BROOK Prolog is by far the most used of various logic programming languages that have been proposed. The reason for this is the existence of very efficient implementations. This tutorial will show in detail how this efficiency is achieved. The first half of this tutorial will concentrate on Prolog compilation. The approach is first to define a Prolog Virtual Machine (PVM), which can be implemented in software, microcode, hardware, or by translation to the language of an existing machine. We will describe in detail the PVM defined by D.H.D. Warren (SRI Technical Note 309) and discuss how its data objects can be represented efficiently. We will also cover issues of compilation of Prolog source programs into efficient PVM programs. ARTIFICIAL INTELLIGENCE AND PROLOG: AN INTRODUCTION TO THEORETICAL ISSUES IN AI WITH PROLOG EXAMPLES GEORGE F. LUGER UNIVERSITY OF NEW MEXICO This tutorial is intended to introduce the important concepts of both Artificial Intelligence and Logic Programming. To accomplish this task, the theoretical issues involved in AI problem solving are presented and discussed. These issues are exemplified with programs written in Prolog that implement the core ideas. Finally, the design of a Prolog interpreter as Resolution Refutation system is presented. The main ideas from AI problem solving that are presented include: 1) An introduction of AI as representation and search. 2) An introduction of the Predicate Calculus as the main representation formalism for Artificial Intelligence. 3) Simple examples of Predicate Calculus representations, including a relational data base. 4) Unification and its role both in Predicate Calculus and Prolog. 5) Recursion, the control mechanism for searching trees and graphs, 6) The design of search strategies, especially depth first, breadth first and best first or "heuristic" techniques, and 7) The Production System and its use both for organizing search in a Prolog data base, as well as the basic data structure for "rule based" Expert Systems. The above topics are presented with simple Prolog program implementations, including a Production System code for demonstrating search strategies. The final topic presented is an analysis of the Prolog interpreter and an analysis of this approach to the more general issue of logic programming. Resolution is considered as an inference strategy and its use in a refutation system for "answer extraction" is presented. More general issues in AI problem solving, such as the relation of "logic" to "functional" programming are also discussed. PARALLELISM IN LOGIC PROGRAMMING JOHN CONERY UNIVERSITY OF OREGON The fields of parallel processing and logic programming have independently attracted great interest among computing professionals recently, and there is currently considerable activity at the interface, i.e. in applying the concepts of parallel computing to logic programming and, more specifically yet, to Prolog. The application of parallelism to Logic Programming takes two basic but related directions. The first involves leaving the semantics of sequential programming, say ordinary Prolog, as intact as possible, and uses parallelism, hidden from the programmer, to improve execution speed. This has traditionally been a difficult problem requiring very intelligent compilers. It may be an easier problem with logic programming since parallelism is not artificially made sequential, as with many applications expressed in procedural languages. The second direction involves adding new parallel programming primitives to Logic Programming to allow the programmer to explicitly express the parallelism in an application. This tutorial will assume a basic knowledge of Logic Programming, but will describe current research in parallel computer architectures, and will survey many of the new parallel machines, including shared-memory architectures (RP3, for example) and non-shared-memory architectures (hypercube machines, for example). The tutorial will then describe many of the current proposals for parallelism in Logic Programming, including those that allow the programmer to express the parallelism and those that hide the parallelism from the programmer. Included will be such proposals as Concurrent Prolog, Parlog, Guarded Horn Clauses (GHC), and Delta-Prolog. An attempt will be made to partially evaluate many of these proposals for parallelism in Logic Programming, both from a pragmatic architectural viewpoint as well as from a semantic viewpoint. Conference Chairperson Gary Lindstrom, University of Utah Program Chairperson Robert M. Keller, University of Utah Local Arrangements Chairperson Thomas C. Henderson, University of Utah Tutorials Chairperson George Luger, University of New Mexico Exhibits Chairperson Ross Overbeek, Argonne National Lab. Program Committee Francois Bancilhon, MCC John Conery, U. of Oregon Al Despain, U.C. Berkeley Herve Gallaire, ECRC, Munich Seif Haridi, SICS, Stockholm Lynette Hirschman, SDC Peter Kogge, IBM, Owego William Kornfeld, Quintus Systems Gary Lindstrom, University of Utah George Luger, University of New Mexico Rikio Onai, ICOT/NTT, Tokyo Ross Overbeek, Argonne National Lab. Mark Stickel, SRI International Sten Ake Tarnlund, Uppsala University ------------------------------ End of AIList Digest ******************** From csnet_gateway Thu Sep 18 06:51:39 1986 Date: Thu, 18 Sep 86 06:51:32 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #182 Status: R AIList Digest Wednesday, 17 Sep 1986 Volume 4 : Issue 182 Today's Topics: Conference - ISMIS'86 program ---------------------------------------------------------------------- Date: Thu, 11 Sep 86 17:03 EST From: ZEMANKOVA%tennessee.csnet@CSNET-RELAY.ARPA Subject: Conference - ISMIS'86 program PRELIMINARY PROGRAM INTERNATIONAL SYMPOSIUM ON METHODOLOGIES FOR INTELLIGENT SYSTEMS October 22 - 25, 1986 Hilton Hotel Knoxville, Tennessee Sponsored by * ACM Special Interest Group on Artificial Intelligence in cooperation with * University of Tennessee at Knoxville * The Data Systems Research and Development Program of Martin Marietta Energy Systems, and Oak Ridge National Laboratory * University of North Carolina at Charlotte and hosted by * The Procter and Gamble Company CHAIRPERSONS Zbigniew W. Ras (UTK and UNCC) Maria Zemankova (UTK and UNCC) SYMPOSIUM COORDINATOR J. Robin B. Cockett (UTK) ORGANIZING COMMITTEE S. Chen (IUPUI) M. Emrich (ORNL) G. Epstein (UNCC & Indiana) K. O'Kane (UTK) J. Poore (Georgia Tech.& UTK) R. Yager (Iona) PROGRAM COMMITTEE P. Andrews (Carnegie-Mellon) J. Bourne (Vanderbilt) M. Fitting (CUNY) B. Gaines (Calgary, Canada) M. Gupta (Saskatchewan, Canada) M. Karpinski (Bonn, West Germany) E. Knuth (Budapest, Hungary) S. Kundu (LSU) W. Marek (Kentucky) R. Michalski (Illinois-Urbana) C. Negoita (CUNY) R. Nelson (Case Western Reserve) Z. Pawlak (Warsaw, Poland) A. Pettorossi (Rome, Italy) E. Sandewall (Linkoping, Sweden) G. Shafer (Kansas) M. Shaw (Calgary, Canada) J. Tou (Florida) PURPOSE OF THE SYMPOSIUM This Symposium is intended to attract researchers who are actively engaged both in theoretical and practical aspects of intelligent systems. The goal is to provide a platform for a useful exchange between theoreticians and practitioners, and to foster the crossfertilization of ideas in the following areas: * Expert Systems * Knowledge Representation * Logic for Artificial Intelligence * Learning and Adaptive Systems * Intelligent Databases * Approximate Reasoning There will be an exhibit of A.I. hardware and software and of A.I. literature. Symposium Proceedings will be published by ACM Press. ISMIS 86 Symposium Schedule Tuesday, October 21, 1986 ========================= 6:00 pm - 9:00 pm Symposium Registration 7:00 pm - 9:00 pm Reception (Cash Bar) 6:00 pm - 9:00 pm Exhibits Wednesday, October 22, 1986 =========================== 8:00 am - 12:00 am Symposium Registration ISMIS'86 Opening Session 9:00 am - 9:20 am Session 1: Expert Systems I1: Invited Papers Chair: M. Emrich (ORNL) 9:20am - 10:05am "Recent Developments in Expert Systems" B. Buchanan (Stanford Univ.) 10:05am - 10:50am "Generic Tasks in Artificial Intelligence and Mentalese" B. Chandrasekaran (Ohio State Univ.) A1: Contributed Papers Chair: R. Cockett (UT Knoxville) 11:15am - 11:40am "The Frame-Definition Language for Customizing the Raffaello Structure-Editor in Host Expert Systems" E. Nissan (Ben-Gurion, Israel) 11:40am - 12:05am "Knowledge Base Organization in Expert Systems" S. Frediani, L. Saitta (Torino, Italy) 12:05am - 12:30pm "NESS: A Coupled Simulation Expert System" K. Kawamura, G. Beale, J. Rodriguez-Moscoso, B.J. Hsieh, S. Padalkar (Vanderbilt) B1: Contributed Papers Chair: J. Bourne (Vanderbilt) 11:15am - 11:40am "Design of an Expert System for Utilization Research" A. Zvieli, S.K. MacGregor, J.Z. Shapiro (LSU) 11:40am - 12:05am "An Expert System for Dynamic Scheduling" S. Floyd, D. Ford (Huntsville, Alabama) 12:05am - 12:30pm "Beginners' Strategies in Example Based Expert Systems" T. Whalen, B. Schott (Atlanta, Georgia) 12:30 pm - 2:00 pm Exhibits Session 2: Intelligent Databases I2: Invited Papers Chair: W. Marek (UK Lexington) 2:00pm - 2:45pm "Using Knowledge Representation for the Development of Interactive Information Systems" J. Mylopoulos (Toronto, Canada) 2:45pm - 3:30pm "Acquisition of Knowledge from Data" G. Wiederhold (Stanford Univ.) A2: Contributed Papers Chair: S. Kundu (LSU) 3:50pm - 4:15pm "A Decidable Query Answering Algorithm for Circumscriptive Theories" T. Przymusinski (El Paso, Texas) 4:15pm - 4:40pm "Fuzzy Knowledge Engineering Techniques in Scientific Document Classification" R. Lopez de Mantaras (Barcelona, Spain) 4:40pm - 5:05pm "A Semantic and Logical Front-end to a Database System" M. Rajinikanth, P.K. Bose (Texas Instruments, Dallas) 5:05pm - 5:30pm "A Knowledge-Based Approach to Online Document Retrieval System Design" G. Biswas, J.C. Bezdek, R.L. Oakman (Columbia, S.C.) 5:30pm - 5:55pm "Towards an Intelligent and Personalized Information Retrieval System" S.Myaeng, R.R. Korfhage (Southern Methodist, Texas) 6:00 pm - 7:30 pm Exhibits 7:30 pm - 10:00 pm Dinner Theatre Karel Capek, R.U.R. Thursday, October 23, 1986 ========================== Session 3: Approximate Reasoning I3: Invited Papers Chair: M. Zemankova (UT Knoxville) 9:00am - 9:45am "Inductive Models under Uncertainty" P. Cheeseman (NASA AMES and SRI) 9:45am - 10:30am "The Concept of Generalized Assignment Statement and its Application to Knowledge Representation in Fuzzy Logic" L.A. Zadeh (Berkeley) A3: Contributed Papers Chair: B. Bouchon (Paris, France) 10:50am - 11:15am "Expert System on a Chip: An Engine for Real-Time Approximate Reasoning" M. Togai (Rockwell International), H. Watanabe (AT&T Bell Lab, Holmdel) 11:15am - 11:40am "Selecting Expert System Frameworks within the Bayesian Theory" S.W. Norton (PAR Government Systems Co., New Hartford) 11:40am - 12:05pm "Inference Propagation in Emitter, System Hierarchies" T. Sudkamp (Wright State) 12:05pm - 12:30pm "Estimation of Minimax Values" P. Purdom (Indiana), C.H. Tzeng (Ball State Univ.) B3: Contributed Papers Chair: E. Nissan (Ben-Gurion, Israel) 10:50am - 11:15am "Aggregating Criteria with Quantifiers" R.R. Yager (Iona College) 11:15am - 11:40am "Approximating Sets with Equivalence Relations" W. Marek (Kentucky), H. Rasiowa (Warsaw, Poland) 11:40am - 12:05pm "Evidential Logic and Dempster-Shafer Theory" S. Chen (UNC-Charlotte) 12:05pm - 12:30pm "Propagating Belief Functions with Local Computations" P.P. Shenoy, G. Shafer (Lawrence, Kansas) 12:30 pm - 2:00 pm Exhibits Session 4: Logics for Artificial Intelligence I4: Invited Papers Chair: M. Fitting (CUNY) 2:00pm - 2:45pm "Automated Theorem Proving: Mapping Logic into A.I." D.W. Loveland (Duke Univ.) 2:45pm - 3:30pm "Extensions to Functional Programming in Scheme" D.A. Plaisted, J. W. Curry (UNC Chapel Hill) A4: Contributed Papers Chair: G. Epstein (UNC Charlotte) 3:50pm - 4:15pm "Logic Programming Semantics using a Compact Data Structure" M. Fitting (CUNY) 4:15pm - 4:40pm "On the Relationship between Autoepistemic Logic and Parallel Circumscription" M. Gelfond, H. Przymusinska (El Paso, Texas) 4:40pm - 5:05pm "A Preliminary Excursion Into Step-Logics" J. Drapkin, D. Perlis (College Park, Maryland) 5:05pm - 5:30pm "Tree Resolution and Generalized Semantic Tree" S. Kundu (LSU) 5:30pm - 5:55pm "An Inference Model for Inheritance Hierarchies with Exceptions" K. Whitebread (Honeywell, Minneapolis) 6:00 pm - 7:30 pm Exhibits 7:30 pm - 9:30 pm Symposium Banquet Keynote Speaker: Brian Gaines (Calgary, Canada) Friday, October 24, 1986 ======================== Session 5: Learning and Adaptive Systems I5: Invited Papers Chair: Z. Ras (UT Knoxville) 8:45am - 9:30am "Analogical Reasoning in Planning and Decision Making" J. Carbonell (Carnegie-Mellon Univ.) 9:30am - 10:15am "Emerging Principles in Machine Learning" R. Michalski (Univ. of Illinois at Urbana) A5: Contributed Papers Chair: D. Perlis (Maryland) 10:35am - 11:00am "Memory Length as a Feedback Parameter in Learning Systems" G. Epstein (UNC-Charlotte) 11:00am - 11:25am "Experimenting and Theorizing in Theory Formation" B. Koehn, J.M. Zytkow (Wichita State) 11:25am - 11:50am "On Learning and Evaluation of Decision Rules in the Context of Rough Sets" S.K.M. Wong, W. Ziarko (Regina, Canada) 11:50am - 12:15pm "Taxonomic Ambiguities in Category Variations Needed to Support Machine Conceptualization" L.J. Mazlack (Berkeley) 12:15pm - 12:40pm "A Model for Self-Adaptation in a Robot Colony" T.V.D.Kumar, N. Parameswaran (Madras, India) 12:45 pm - 2:00 pm Symposium Luncheon Keynote Speaker: Joseph Deken (NSF) "Viable Inference Systems" Session 6: Knowledge Representation I6: Invited Papers Chair: S. Chen (UNC Charlotte) 2:15pm - 3:00pm "Self-Improvement in Problem-Solving" R.B. Banerji (St. Joseph's Univ.) 3:00pm - 3:45pm "Logical Foundations for Knowledge Representation in Intelligent Systems" B.R. Gaines (Calgary, Canada) A6: Contributed Papers Chair: M. Togai (Rockwell International) 4:00pm - 4:25pm "Simulations and Symbolic Explanations" D.H. Helman, J.L. Bennett, A.W. Foster (Case Western Reserve) 4:25pm - 4:50pm "Notes on Conceptual Representations" E. Knuth, L. Hannak, A. Hernadi (Budapest, Hungary) 4:50pm - 5:15pm "Spaceprobe: A System for Representing Complex Knowledge" J. Dinsmore (Carbondale, Ill) 5:15pm - 5:40pm "Challenges in Applying Artificial Intelligence Methodologies to Military Operations" L.F. Arrowood, M.L. Emrich, M.R. Hilliard, H.L. Hwang (Oak Ridge National Lab.) B6: Contributed Papers Chair: L. de Mantaras (Barcelona, Spain) 4:00pm - 4:25pm "Knowledge-Based Processing/Interpretation of Oceanographic Satellite Data" M.G. Thomason, R.E. Blake (UTK), M. Lybanon (NTSL) 4:25pm - 4:50pm "A Framework for Knowledge Representation and use in Pattern Analysis" F. Bergadano, A. Giordana (Torino, Italy) 4:50pm - 5:15pm "Algebraic Properties of Knowledge Representation Systems" J.W. Grzymala-Busse (Lawrence, Kansas) 5:15pm - 5:40pm "Prime Rule-based Methodologies Give Inadequate Control" J.R.B. Cockett, J. Herrera (UTK) ISMIS'86 Closing Session 5:45pm - 6:00pm Saturday, October 25, 1986 ========================== 9:00 am - 12:30 pm Colloquia (parallel sessions) 1:30 pm - 7:30 pm Trip to the Smoky Mountains SYMPOSIUM FEES Advance Symposium Registration Received by September 15, 1986 Member of ACM $220.00 Non-member $250.00 Student* $ 30.00 Late or On-Site Registration Member of ACM $265.00 Non-member $295.00 Student* $ 40.00 Additional Tickets Reception $ 5.00 Dinner Theatre $ 25.00 Symposium Banquet $ 25.00 Symposium Luncheon $ 10.00 Trip to Smoky Mountains $ 25.00 Symposium registration fee includes the Proceedings (available at the Symposium), continental breakfasts, reception, dinner theatre, symposium banquet, symposium luncheon, coffee breaks. * Student registration includes only coffee breaks. Students registration limited, hence students should register early. ACCOMMODATIONS: A block of rooms has been reserved for the symposium at the Hilton Hotel. The ISMIS 86 rate for a single occupancy is $47.00 and double occupancy $55.00. To reserve your room, contact the Hilton Hotel, 501 Church Avenue, S.W., Knoxville, TN 37902-2591, telephone 615-523-2300 by September 30, 1986. The Hilton Hotel will continue to accept reservations after this date on a space availability basis at the ISMIS 86 rates. However, you are strongly encouraged to make your reservations by the cutoff date of September 30. Reservation must be accompanied by a deposit of one night's room rental. TRANSPORTATION: The Hilton Hotel provides a free limousine service from and to the airport. If arriving by your vehicle, all overnight guests receive free parking. SPECIAL AIRFARE RATES: DELTA Airlines has been designated as the official carrier for the Symposium. Attendees arranging flights with DELTA will receive a 35% discount off the regular coach fare to Knoxville. To take advantage of this speical rate call (toll-free) 1-800-241-6760, referring to FILE #J0170. This number is staffed from 8:00 a.m. to 8:00 p.m. EDT, seven days per week. GENERAL INFORMATION: Knoxville is located in East Tennessee, the area that is noted for its abundant water reservoirs, rivers, mountains, hardwood forests and wildlife refuges. The Great Smoky Mountains National Park, the Cumberland Mountains, the resort city of Gatlinburg, and the Oak Ridge Museum of Science and Energy are all within an hours drive from the downtown area. The Fall season offers spectacular views of radiant colors within the city and the surrounding contryside. Interstates 40 and 75 provide access into Knoxville. REGISTRATION FORM: For the registration form, please write to UTK Departments of Conferences 2014 Lake Avenue Knoxville, TN 37996-3910 FURTHER INFORMATION: Further information can be obtained from: Zbigniew W. Ras Maria Zemankova Dept. of Computer Science Dept. of Computer Science University of North Carolina University of Tennessee Charlotte, NC 28223 Knoxville, TN 37996-1301 (704) 597-4567 (615) 974-5067 ras%unccvax@mcnc.CSNET zemankova@utenn.CSNET ------------------------------ End of AIList Digest ******************** From csnet_gateway Fri Sep 19 19:08:51 1986 Date: Fri, 19 Sep 86 19:08:43 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #183 Status: R AIList Digest Wednesday, 17 Sep 1986 Volume 4 : Issue 183 Today's Topics: Queries - Space/Military Expert Systems & Communications/Control ES & Structured Analysis References & NuBus-to-VME Adapter & Robotic Cutting Arm & Schematics Drafter & Mechanical Engineering ES & OPS5 & IJCAI-87 Net Address & Common Lisp Flavors & Looping in Belief Revision System & 2-D Math Editor ---------------------------------------------------------------------- Date: 16 Aug 86 15:39:32 GMT From: mcvax!ukc!reading!brueer!ckennedy@seismo.css.gov (C.M.Kennedy ) Subject: CONTACT REQUIRED: SPACE OR MILITARY EXPERT SYSTEMS CONTACT REQUIRED: SPACE OR MILITARY EXPERT SYSTEMS I wish to contact someone (reasonably senior) who has worked on an expert system in one of the following areas: 1. Space technology - monitoring, control, planning 2. Military Science - of particular interest is: - prediction, e.g. modelling behaviour of states or terrorist organisations and making predictions based on available knowledge - interpretation of sensor data. i.e. integrating raw data from multiple sensors and giving a high-level "user- friendly" interpretation of what is going on. I wish to obtain the following information: 1. Postal address and telephone number (along with email address). If possible: Times of day (and days of the week) when telephone contact is convenient. 2. Details of how to obtain the following documentation (or better still a direct mailing of it if this is convenient): - TECHNICAL papers describing the architecture, knowledge representation, inference engine, tools, language, machine etc. - papers giving the precise REQUIREMENTS of the system. If this is not possible, a short summary will do. 3. Was the project successful? Were all the original requirements satisfied? Has the system been used successfuly in an operational environment? 4. What were the problems encountered and what has been learned from the project? I would also be interested to hear from someone who has done RESEARCH on any of the above (or knows of someone who has). Catriona Kennedy Mail address: ckennedy@ee.brunel.ac.uk ------------------------------ Date: Mon 25 Aug 86 12:18:41-EDT From: CAROZZONI@RADC-TOPS20.ARPA Subject: Cooperative Expert System The Decision Aids Section at Rome Air Development Center is performing an in-house study to establish a technical baseline in support of an upcoming (FY 87) procurement effort related to the design of a "cooperative" expert system - i.e., one which supports both communication and more extensive, knowledge-based cooperation between existing systems. We are particularly interested in hearing about any work related to expert system design, distributed AI, and models for communication and cooperation that may be relevant to this effort. Please respond by net to Hirshfield@RADC-multics, or write to Hirshfield at RADC/COAD, Griffiss Air Force Base, NY 13441. ------------------------------ Date: Wed 10 Sep 86 15:49:36-CDT From: Werner Uhrig Subject: Communications Expert System - does anyone know more ? [ from InfoWorld, Sep 8, page 16 ] COMMUNICATIONS PROGRAM TO HELP NOVICES, EXPERTS Smyran, Ga - A communications software pulisher said it wil sell an on-line expert system that helps computer users solve data communications problems and work out idiosyncracies in the interaction of popular communications hardware and software. Line Expert, which will sell for $49.95 when it is released October 1, will ask users questions about their particular configuration and suggest solutions, according to Nat Atwell, director of marketing for publisher Concept Development Systems. .......... ------------------------------ Date: Mon, 8 Sep 86 22:25:30 cdt From: Esmail Bonakdarian Subject: Expert Systems and Data Communication I am working on my M.S. thesis which deals with the use of Expert Systems in the area of Data Communications (e.g. help diagnose sources of communication problems, help to "configure" components [DTE's and DCE's] correctly, etc). I am curious to find out what knowledge based systems (if any) exist that deal with this problem domain. I would very much appreciate any pointers to literature or persons doing work in this area. Thanks, Esmail ------------------------------ Date: Wed, 20 Aug 86 9:45:15 EDT From: Marty Hall Subject: Wanted: References on Structured Analysis Inadequacies We are looking for references that point out some of the inadequacies of Structured Analysis methods (ala Yourdon, for instance) in a Software Development Process for AI software. We have a couple of references vouching for the utility of Rapid Prototyping and Exploratory Programming (thanks, by the way, for those who pointed me to some of these references), but not explicitly contrasting this with the more traditional Structured Design/Analysis methods. These references are needed by our AI group for a "Convince the Software Managers" session. :-) Any help greatly appreciated! - Marty Hall Arpa: hall@hopkins AI and Simulation Dept, MP E-315 UUCP: seismo!umcp-cs!aplcen!jhunix!ins_amrh Martin Marietta Baltimore Aerospace 103 Chesapeake Park Plaza Baltimore, MD 21220 (301) 682-0917 ------------------------------ Date: 15 Aug 86 14:32:00 GMT From: pyrnj!mirror!datacube!berger@CAIP.RUTGERS.EDU Subject: NuBus to VME adapter? I figured this would be as good a place as any for the following question: Anyone know of a NuBus to VMEbus adapter? Something to allow VMEbus boards to plug into a NuBus? We want to beable to connect our Image Processing boards into things like the TI explorer and LMI machines. Bob Berger Datacube Inc. 4 Dearborn Rd. Peabody, Ma 01960 617-535-6644 ihnp4!datacube!berger {seismo,cbosgd,cuae2,mit-eddie}!mirror!datacube!berger ------------------------------ Date: Thu, 11 Sep 86 10:44 MST From: McGuire@HIS-PHOENIX-MULTICS.ARPA Subject: robotics query: cutting arm Could anyone give possible sources for a robotic arm, to be attached to a CAD/CAM system(such as Auto-Cad), driven by a micro, such as a PC/AT? This arm would be used to cut stencils, maximum 3 feet diameter, so it would have to be very strong or complex. Canadian sources preferred. Thanks. M.McGuire, Calgary, Alberta. ------------------------------ Date: 19 Aug 86 12:41:39 edt From: Raul Valdes-Perez Subject: schematics drafting request I have designed and programmed a non-rule-based KBES that drafts the schematic of a digital circuit (actually only the placement part). To have an objective measure of the ability of this program, I would like to compare its output with that of any other (perhaps algorithmic) schematics drafter. I expect that a large CAD circuit design package would have something like this. Can anyone help me obtain access to such a drafter? (Please note that this has little to do with a schematic *entry* program, nor with a VLSI *layout* program. Thanks in advance. Raul E. Valdes-Perez or (valdes@mit-htvax.arpa) MIT AI Lab, Room 833 545 Technology Square Cambridge, MA 02139 ------------------------------ Date: Wed, 3 Sep 86 08:16 CDT From: Bennett@HI-MULTICS.ARPA Subject: Looking for Expert Systems for Mechanical Engineering A friend of mine is looking for pointers to work done in Expert Systems for Mechanical Engineering --- specifically in the area of Mechanical Design. If anyone has any information that would help please send it directly to me as Bennett at HI-Multics. Bonnie Bennett (612)782-7381 ------------------------------ Date: Mon, 25 Aug 86 22:54 EDT From: EDMUNDSY%northeastern.edu@CSNET-RELAY.ARPA Subject: Any OPS5 in PC ? Does anyone know whether there is any OPS5 software package availiable in PC? I would like to know where I can find it. Thanks!!! ------------------------------ Date: 27 Aug 86 10:49:18 GMT From: ulysses!mhuxr!aluxp!prieto@ucbvax.Berkeley.EDU (PRIETO) Subject: ATT 3B2/400, 3B5 CMU - OPS5 What are the OPS5 requirements to be used in small machines like 3B2/400, 3B5 vs. VAX 11/780? Storage, memory, etc. Is there OPS5 software executing in these types of machines? Can software development for an expert system application be done in the smaller machines or is a VAX needed? aluxp!prieto (215)770-3285 ps. I am interested in getting OPS 5 - where could I obtain it? ------------------------------ Date: 13 Aug 86 04:48:39 GMT From: ucbcad!nike!lll-crg!micropro!ptsfa!jeg@ucbvax.berkeley.edu (John Girard) Subject: IJCAI-87 ... usenet contacts I am looking for a usenet or usenet compatible connection by which I can inquire about the IJCAI program, ground rules and deadlines. Please respond to [ihnp4,dual,pyramid,cbosgd,bellcore,qantel]ptsfa!jeg John Girard USA: 415-823-1961 415-449-5745 ------------------------------ Date: Mon, 8 Sep 86 18:33:09 -0100 From: mcvax!csinn!solvay@seismo.CSS.GOV (Jean Philippe Solvay) Subject: flavors and Common Lisp Hi Kenneth, Do yo know if there is any implementation of flavors in Common Lisp currently available (public domain, if possible)? Thanks in advance, Jean-Philippe Solvay. inria!csinn!solvay@mcvax.UUCP ------------------------------ Date: Mon, 08 Sep 86 16:48:15 -0800 From: Don Rose Subject: TMS, DDB and infinite loops Does anyone know whether the standard algorithms for belief revision (e.g. dependency-directed backtracking in TMS-like systems) are guaranteed to halt? That is, is it possible for certain belief networks to be arranged such that no set of mutually consistent beliefs can be found (without outside influence)? --Donald Rose drose@ics.uci.edu ICS Dept Irvine CA 92717 ------------------------------ Date: 0 0 00:00:00 PDT From: "LLLASD::GARBARINI" Reply-to: "LLLASD::GARBARINI" Subject: Availability of interactive 2-d math editing interfaces... I am working with a number of other people on a project called Automatic Programming for Physics. The goal is to build an AI based automatic programming system to aid scientist in the building of numerical simulations of physical systems. In the user interface to the system we would like to have interactive editing of mathematical expressions in two-dimensional form. It seems a number of people have recently made much progress in this area. (See C. Smith and N. Soiffer, "MathScribe: A User Interface for Computer Algebra Systems," Conference Proceedings of Symsac 86, (July, 1986) and B. Leong, "Iris: Design of a User Interface Program for Symbolic Algebra," Proc. 1986 ACM-SIGSAM Symposium on Symbolic and Algebraic Manipulation, July 1986.) Not wishing to reinvent the wheel, I'd appreciate receiving information regarding the availability of any such interface. Joe P. Garbarini Jr. Lawrence Livermore National Lab P. O. Box 808 , L-308 7000 East Avenue Livermore Ca. , 94550 (415)-423-2808 arpanet address: GARBARINI%LLLASD.DECNET@LLL-CRG.ARPA ------------------------------ End of AIList Digest ******************** From csnet_gateway Sat Sep 20 00:54:04 1986 Date: Sat, 20 Sep 86 00:53:57 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #184 Status: R AIList Digest Thursday, 18 Sep 1986 Volume 4 : Issue 184 Today's Topics: Correction - Conference on Office Information Systems, AI Tools - Interlisp vs. C, Queries - NL Grammar & Unix Software, Education - AI Schools, AI Tools - Turbo Prolog ---------------------------------------------------------------------- Date: Wed, 17 Sep 86 14:13:52 cdt From: preece%ccvaxa@gswd-vms.ARPA (Scott E. Preece) Subject: Correction - Conference on Office Information Sys > From: rba@petrus.bellcore.com (Robert B. Allen) > Subject: Conference on Office Information Systems - Brown U. > > > ACM CONFERENCE ON OFFICE INFORMATION SYSTEMS > October 6-8, 1968, Providence, R.I. ^ Gee, I didn't join the ACM until 1970, but I didn't think they had invented "Office Information Systems" then... -- scott preece gould/csd - urbana uucp: ihnp4!uiucdcs!ccvaxa!preece arpa: preece@gswd-vms ------------------------------ Date: 16 Sep 86 13:07 EDT From: Denber.wbst@Xerox.COM Subject: Re: Reimplementing in C "Such things are very awkward, if not impossible to express in the typical AI languages" Well, maybe I've been using an atypical AI language, but Interlisp-D has all that stuff - byte I/O, streams, timers, whatever. It's real e-z to use. Check it out. - Michel ------------------------------ Date: Thu, 14 Aug 86 11:58 EDT From: EDMUNDSY%northeastern.edu@CSNET-RELAY.ARPA Subject: Looking for Production Rules for English Grammar Does anyone know where can I find the information (or existed results) of transforming English (or a simplified subset) grammar into production rules of regular grammar, context-free or context sensitive grammar. For example, Sentences --> Noun Verb Noun etc. If anyone gets any information on that, I would appreciate if you can leave me a pointer for those information. Thanks!! I can be contacted by any of the following means: NET: EDMUNDSY@NORTHEASTERN.EDU ADD: Sy, Bon-Kiem Northeastern University Dept. ECE DA 409 Boston, MA 02115 Phone: (617)-437-5055 Bon Sy ------------------------------ Date: Mon, 8 Sep 86 18:32:06 edt From: brant@linc.cis.upenn.edu (Brant A. Cheikes) Subject: Unix Consultant references? I'm looking for the most recent reports by the group working on the Unix Consultant project at UC Berkeley. Does anybody know what that is, and is there a network address to which report requests can be sent? The ref I was given was UCB report CSD 87/303, but I'm not sure if it's available or even recent. Any information in this vein would be appreciated. ------------------------------ Date: 12 Sep 86 00:32:34 GMT From: micropro!ptsfa!jeg@lll-crg.arpa (John Girard) Subject: AI tools/products in UNIX Greetings, I am looking for any information I can get on Artificial Intelligence tools and products in the UNIX environment. I will compile and publish the results in net.ai. Please help me out with any of the following: versions of LISP and PROLOG running in unix expert system shells available in unix expert system and natural language products that have been developed in the unix environment, both available now and in R&D, especially ones that relate to unix problem domains (sys admin, security). Reply to: John Girard 415-823-1961 [ihnp4,dual,cbosgd,nike,qantel,bellcore]!ptsfa!jeg P.S. Very interested in things that run on less horsepower than a SUN. ------------------------------ Date: Sat, 13 Sep 86 13:44:47 pdt From: ucsbcsl!uncle@ucbvax.Berkeley.EDU Subject: request for core nl system code We are looking for a core nl system which we can tailor and extend. There is as yet little comp.ling activity at UCSB, so we have no local sources. We are interested in developing a system which can be used in foreign language education, hence we would need a system in which the "syntactic components" are such that we could incrementally mung the system into speaking german or french or russian without having to redesign the system. my knowledge in this area is fuzzy (not 'Fuzzy(tm)' etc, just fuzzy!) . I have read a little about systems such as the Phran component of the Wilensky et al project called unix-consultant, and i understand that the approach taken there is susceptible to generalization to other languages by entering a new data-base of pattern-action pairs (i.e. an EXACT parse of a syntactically admissable sentence is not required) Unfortunately, Berekeley CS is not currently giving access to components of that system. Does anyone have pointers to available code for systems that fall into that part of the syntax-semantics spectrum? Is it, in fact, reasonable for us to seek such a system as a tool, or are we better advised to start with car and cdr ???? ------------------------------ Date: 19 Aug 86 19:29:25 GMT From: decvax!dartvax!kapil@ucbvax.Berkeley.EDU (Kapil Khetan) Subject: Where can one do an off-campus Ph.D. in AI/ES After graduating from Dartmouth, with an MS in Computer & Information Science, I have been residing and working in New York City. I am interested in continuing education and think Expert Systems is a nice field to learn more about. I took a ten week course in which we dabbled in Prolog and M1. If any of you know of a college in the area (Columbia, NYU, PACE) which has something like it, or any other college anywhere else which has an off-campus program, please hit the 'r' key. Thank-you. Kapil Khetan Chemical Bank, 55 Water St., New York, NY 10041 ------------------------------ Date: 25 Aug 86 18:27:08 GMT From: ihnp4!gargoyle!sphinx!bri5@ucbvax.Berkeley.EDU (Eric Brill) Subject: Grad Schools Hello. I am planning on entering graduate school next year. I was wondering what schools are considered the best in Artificial Intelligence (specifically in language comprehension and learning). I would be especially interested in your opinions as to which schools would be considered the top 10. Thank you very much. Eric Brill ps, if there is anybody else out there interested in the above, send me mail, and I will forward all interesting replies. ------------------------------ Date: Fri, 12 Sep 86 15:35 CDT From: PADIN%FNALB.BITNET@WISCVM.WISC.EDU Subject: ADVICE ON ENTERING THE AI COMMUNITY As a newcomer to the AI arena, I am compelled to ask some fundamentally novice (and,as such,sometimes ridiculous) sounding questions. Nonetheless, here goes. If one were to attempt to enter the AI field, what are the basic requirements; what are some special requirements? With a BS im physics, is further schooling mandatory? Are there particular schools which I should consider or ones I should avoid? Are there books which I MUST read!!? As a 29 year old with a Math and Physics background, am I hopelessly over-the-hill for such musings to become a reality? Are there questions which I should be asking? If you care to answer in private I can be reached at: PADIN@FNALB.BITNET ------------------------------ Date: 2 Sep 86 21:44:00 GMT From: pyrnj!mirror!prism!mattj@CAIP.RUTGERS.EDU Subject: Re: Grad Schools Eric Brill: Here is my own personal ranking of general AI programs: Stanford MIT Carnegie-Mellon UIllinois@Urbana URochester Also good: UMaryland, Johns Hopkins, UMass@Amherst, ... can't think now. [...] - Matthew Jensen ------------------------------ Date: 6 Sep 86 02:52:06 GMT From: ubc-vision!ubc-cs!andrews@UW-BEAVER.ARPA (Jamie Andrews) Subject: Re: Grad Schools (Rochester?) I've heard that Rochester has quite a good AI / logic programming program, and it definitely has some good people... but can anyone tell me what it's like to LIVE in Rochester? Or is the campus far enough from Rochester that it doesn't matter? Please respond (r or R) to me rather than to the net. Adv(merci)ance, --Jamie. ...!seismo!ubc-vision!ubc-cs!andrews "Hundred million bottles washed up on the shore" ------------------------------ Date: 3 Sep 86 21:20:31 GMT From: mcvax!prlb2!lln-cs!pv@seismo.css.gov (Patrick Vandamme) Subject: Bug in Turbo Prolog I am testing the `famous' Turbo Prolog Software and, after all the good things that I heard about it, I was very surprised at having problems with the first large program I tried. I give here this program. It finds all the relations between a person and his family. But for some people, it answers with a lot of strange characters. I think there must be a dangling pointer somewhere. Note that this happens only with large programs ! Have someone the same result ? (for the stange characters, try with "veronique"). ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /* +-----------------------------------------------------+ | Programme de gestion d'une base de donnees | | de relations familiales. | +-----------------------------------------------------+ P. Vandamme - Unite Info - UCL - Aout 1986 */ [Deleted due to length. See following message for an explanation of the problem. -- KIL] ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ -- Patrick Vandamme Unite d'Informatique UUCP : (prlb2)lln-cs!pv Universite Catholique de Louvain Phone: +32 10 43 24 15 Place Sainte-Barbe, 2 Telex: 59037 UCL B B-1348 Louvain-La-Neuve Eurokom: Patrick Vandamme UCL Belgium Fax : +32 10 41 56 47 ------------------------------ Date: 8 Sep 86 23:08:05 GMT From: clyde!cbatt!cbuxc!cbuxb!cbrma!clh@CAIP.RUTGERS.EDU (C.Harting) Subject: Re: Bug in Turbo Prolog I purchased Turbo Prolog Friday night, and immediately tried to compile the GeoBase program on my Tandy 1000 (384K). I could not even create a .OBJ file on my machine, so I compiled it on a 640K AT&T PC6300. Caveat No. 1: large programs need large amounts of memory. I compiled Patrick's "programme de gestion" to disk and it ran flawlessly (I think -- this is my first lesson in French!). BUT when compiled to memory, I got the same errors as Patrick. Caveat No. 2: compile large programs to disk and run standalone. And, Caveat No. 3: leave out as many memory-resident programs as you can stand when booting the machine to run Turbo Prolog. 'Nuff said? =============================================================================== Chris Harting "Many are cold, few are frozen." AT&T Network Systems Columbus, Ohio The Path (?!?): cbosgd!cbrma!clh ------------------------------ Date: 11 Sep 86 18:40:05 GMT From: john@unix.macc.wisc.edu (John Jacobsen) Subject: Re: Re: Bug in Turbo Prolog > Xref: uwmacc net.ai:1941 net.lang.prolog:528 > Summary: How to get around it. I got the "Programme de Gestion de Base de Donnees" to work fine... on an AT with a meg of memory. I think Patrick Vandamme just ran out of memory, cause his code is immaculate. John E. Jacobsen University of Wisconsin -- Madison Academic Computing Center ------------------------------ Date: Tue, 16 Sep 86 17:26 PDT From: jan cornish Subject: Turbo Prolog I've heard some chilling things about Turbo Prolog. Such as 1) The programmer must not only declare each predicate, but also whether each parameter to the predicate (not correct terminology) is input or output. This means you can't write relational predicates like grandfather. 2) The backtracking is not standard. 3) "You can do any thing in Turbo Prolog that you can do in Turbo Pascal" I want to hear from the LP community on Turbo Prolog as to it's ultimate merit. Something beyond the dismissive flames. Thanks in advance, Jan ------------------------------ End of AIList Digest ******************** From csnet_gateway Sat Sep 20 00:53:32 1986 Date: Sat, 20 Sep 86 00:53:26 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #185 Status: R AIList Digest Friday, 19 Sep 1986 Volume 4 : Issue 185 Today's Topics: Query - Connectionist References, Cognitive Psychology - Connectionist Learning, Review - Notes on AAAI '86 ---------------------------------------------------------------------- Date: 21 Aug 86 12:11:25 GMT From: lepine@istg.dec.com@decwrl.dec.com (Normand Lepine 225-6715) Subject: Connectionist references I am interested in learning about the connectionist model and would appreciate any pointers to papers, texts, etc. on the subject. Please mail references to me and I will compile and post a bibliography to the net. Thanks for your help, Normand Lepine uucp: ...!decwrl!cons.dec.com!lepine ARPA: lepine@cons.dec.com lepine%cons.dec.com@decwrl.dec.com (without domain servers) ------------------------------ Date: 22 Aug 86 12:04:30 GMT From: mcvax!ukc!reading!brueer!ckennedy@seismo.css.gov (C.M.Kennedy ) Subject: Re: Connectionist Expert System Learning The following is a list of the useful replies received so far: Date: Wed, 30 Jul 86 8:56:08 BST From: Ronan Reilly Sender: rreilly%euroies@reading.ac.uk Subject: Re: Connectionist Approaches To Expert System Learning Hi, What you're looking for, effectively, are attempts to implement production systems within a connectionist framework. Researchers are making progress, slowly but surely, in that direction. The most recent paper I've come across in thge area is: Touretzky, D. S. & Hinton, G. E. (1985). Symbols among the neurons details of a connectionist inference architecture. In Proceedings IJCAI '85, Los Angeles. I've a copy of this somewhere. So if the IJCAI proceedings don't come to hand, I'll post it onto you. There are two books which are due to be published this year, and they are set to be the standard reference books for the area: Rumelhart, D. E. & McClelland, J. L. (1986). Parallel distributed processing: Explorations in the microstructure of cognition. Vol. 1: Foundations. Cambridge, MA: Bradford Books. Rumelhart, D. E. & McClelland, J. L. (1986). Parallel distributed processing: Explorations in the microstructure of cognition. Vol. 2: Applications. Cambridge, MA: Bradford Books. Another good source of information on the localist school of connectionism is the University of Rochester technical report series. They have one report which lists all their recent connectionist reports. The address to write to is: Computer Science Department The University of Rochester Rochester, NY 14627 USA I've implemented a version of the Rochester ISCON simulator in Salford Lisp on our Prime 750. The simulator is a flexible system for building and testing connectionist models. You're welcome to a copy of it. Salford Lisp is a Maclisp variant. Regards, Ronan ...mcvax!euroies!rreilly Date: Sat, 2 Aug 86 09:33:46 PDT From: Mike Mozer Subject: Re: Connectionist Approaches To Expert System Learning I've just finished a connectionist expert system paper, which I'd be glad to send you if you're interested (need an address, though). Here's the abstract: RAMBOT: A connectionist expert system that learns by example Expert systems seem to be quite the rage in Artificial Intelligence, but getting expert knowledge into these systems is a difficult problem. One solution would be to endow the systems with powerful learning procedures which could discover appropriate behaviors by observing an expert in action. A promising source of such learning procedures can be found in recent work on connectionist networks, that is, massively parallel networks of simple processing elements. In this paper, I discuss a Connectionist expert system that learns to play a simple video game by observing a human player. The game, Robots, is played on a two-dimensional board containing the player and a number of computer-controlled robots. The object of the game is for the player to move around the board in a manner that will force all of the robots to collide with one another before any robot is able to catch the player. The connectionist system learns to associate observed situations on the board with observed moves. It is capable not only of replicating the performance of the human player, but of learning generalizations that apply to novel situations. Mike Mozer mozer@nprdc.arpa Date: Fri, 8 Aug 86 18:53:57 edt From: Tom Frauenhofer Subject: Re: Connectionist Approaches To Expert System Learning Organization: U. of Rochester Computing Center Catriona, I am (slightly) familiar with a thesis by Gary Cotrell of the U of R here that dealt with a connectionist approach to language understanding. I believe he worked closely with a psychologist to figure out how people understand language and words, and then tried to model the behavior in a connectionist framework. You should be able to get a copy of the thesis from the Computer Science Department here. It's not expert systems, but it is fascinating. - Tom Frauenhofer ...!seismo!rochester!ur-tut!tfra >From sandon@ai.wisc.edu Sat Aug 9 17:25:29 1986 Date: Fri, 8 Aug 86 11:38:43 CDT From: Pete Sandon Subject: Connectionist Learning Hi, You may have already received this information, but I will pass it along anyway. Steve Gallant, at Northeastern University, has done some work on using a modified perceptron learning algorithm for expert system knowledge acquisition. He has written a number of tech reports in the last few years. His email address is: sig@northeastern.csnet. His postal address is: Steve Gallant College of Computer Science Boston, MA. 02115 --Pete Sandon ------------------------------ Date: 17 Aug 86 22:08:30 GMT From: ix133@sdcc6.ucsd.EDU (Catherine L. Harris) Subject: Q: How can structure be learned? A: PDP [Excerpted from the NL-KR Digest by Laws@SRI-STRIPE.] [Forwarded from USENET net.nlang] [... The following portion discusses connectionist learning. -- KIL] One Alternative to the Endogenous Structure View Jeffrey Goldberg says (in an immediately preceding article) [in net.nlang -B], > Chomsky has set him self up asking the question: "How can children, > given a finite amount of input, learn a language?" The only answer > could be that children are equipped with a large portion of language to > begin with. If something is innate than it will show up in all > languages (a universal), and if something is unlearnable then it, too, > must be innate (and therefore universal). The important idea behind the nativist and language-modularity hypotheses are that language structure is too complex, time is too short, and the form of the input data (i.e., parent's speech to children) is too degenerate for the target grammar to be learned. Several people (e.g., Steven Pinker of MIT) have bolstered this argument with formal "learnability" analyses: you make an estimate of the power of the learning mechanism, make assumptions about factors in the learning situation (e.g., no negative feedback) and then mathematically prove that a given grammar (a transformational grammar, or a lexical functional grammar, or whatever) is unlearnable. My problem with these analyses -- and with nativist assumptions in general -- is that they aren't considering a type of learning mechanism that may be powerful enough to learn something as complex as a grammar, even under the supposedly impoverished learning environment a child encounters. The mechanism is what Rumelhart and McClelland (of UCSD) call the PDP approach (see their just-released from MIT Press, Parallel Distributed Processing: Explorations in the Microstructure of Cognition). The idea behind PDP (and other connectionist approaches to explaining intelligent behavior) is that input from hundred/thousands/millions of information sources jointly combine to specify a result. A rule-governed system is, according to this approach, best represented not by explicit rules (e.g., a set of productions or rewrite rules) but by a large network of units: input units, internal units, and output units. Given any set of inputs, the whole system iteratively "relaxes" to a stable configuration (e.g., the soap bubble relaxing to a parabola, our visual system finding one stable interpretation of a visual illustion). While many/most people accept the idea that constraint-satisfaction networks may underlie phenomenon like visual perception, they are more reluctant to to see its applications to language processing or language acquisition. There are currently (in the Rumelhart and McClelland work -- and I'm sure you cognitive science buffs have already rushed to your bookstore/library!) two convincing PDP models on language, one on sentence processing (case role assignment) and the other on children's acquisition of past-tense morphology. While no one has yet tried to use this approach to explain syntactic acquisition, I see this as the next step. For people interested in hard empirical, cross-linguistic data that supports a connectionist, non-nativist, approach to acquisition, I recommend *Mechanisms of Language Acquisition*, Brain MacWhinney Ed., in press. I realize I rushed so fast over the explanation of what PDP is that people who haven't heard about it before may be lost. I'd like to see a discussion on this -- perhaps other people can talk about the brand of connectionism they're encountering at their school/research/job and what they think its benefits and limitations are -- in explaining the psycholinguistic facts or just in general. Cathy Harris "Sweating it out on the reaction time floor -- what, when you could be in that ole armchair theo-- ? Never mind; it's only til 1990!" ------------------------------ Date: 21 Aug 86 11:28:53 GMT From: ulysses!mhuxr!mhuxt!houxm!hounx!kort@ucbvax.Berkeley.EDU (B.KORT) Subject: Notes on AAAI '86 Notes on AAAI Barry Kort Abstract The Fifth Annual AAAI Conference on Artificial Intelligence was held August 11-15 at the Philadelphia Civic Center. These notes record the author's personal impressions of the state of AI, and the business prospects for AI technology. The views expressed are those of the author and do not necessarily reflect the perspective or intentions of other individuals or organizations. * * * The American Association for Artificial Intelligence held its Fifth Annual Conference during the week of August 11, 1986, at the Philadelphia Civic Center. Approximately 5000 attendees were treated to the latest results of this fast growing field. An extensive program of tutorials enabled the naive beginner and technical- professional alike to rise to a common baseline of understanding. Research and Science Sessions concentrated on the theoretical underpinnings, while the complementary Engineering Sessions focused on reduction of theory to practice. Dr. Herbert Schorr of IBM delivered the Keynote Address. His message was simple and straightforward: AI is here today, it's real, and it works. The exhibit floor was a sea of high-end workstations, running flashy applications ranging from CAT scan imagery to automated fault diagnosis, to automated reasoning, to 3-D scene animation, to iconographic model-based reasoning. Symbolics, TI, Xerox, Digital, HP, Sun, and other vendors exhibited state of the art hardware, while Intellicorp, Teknowledge, Inference, Carnegie-Mellon Group, and other software houses offered knowledge engineering power tools that make short work of automated reasoning. Knowledge representation schema include the ubiquitous tree, as well as animated iconographic models of dynamic systems. Inductive and deductive reasoning and goal-directed logic appear in the guise of forward and backward chaining algorithms which seek the desired chain of nodes linking premiss to predicted conclusion or hypothesis to observed symptoms. Such schema are especially well adapted to diagnosis of ills, be it human ailment or machine malfunction. Natural Language understanding remains a hard problem, due to the inscrutable ambiguity of most human-generated utterances. Nevertheless, silicon can diagram sentences as well as a precocious fifth grader. In limited domain vocabularies, the semantic content of such diagrammatic representations can be reliably extracted. Robotics and vision remain challenging fields, but advances in parallel architectures may clear the way for notable progress in scene recognition. Qualitative reasoning, model-based reasoning, and reasoning by analogy still require substantial human guidance, perhaps because of the difficulty of implementing the interdomain pattern recognition which humans know as analogy, metaphor, and parable. Interesting philosophical questions abound when AI moves into the fields of automated advisors and agents. Such systems require the introduction of Value Systems, which may or may not conflict with individual preferences for benevolent ethics or hard-nosed business pragmatics. One speaker chose the provocative title, "Can Machines Be Intelligent If They Don't Give a Damn?" We may be on the threshold of Artificial Intelligence, but we have a long way to go before we arrive at Artificial Wisdom. Nevertheless, some progress is being made in reducing to practice such esoteric concepts as Theories of Equity and Justice, leading to the possibility of unbiased Jurisprudence. AI goes hand in hand with Theories of Learning and Instruction, and the field appears to be paying dividends in the art and practice of knowledge exchange, following the  strategy first suggested by Socrates some 2500 years ago. The dialogue format abounds, and mixed initiative dialogues seem to capture the essence of mutual teaching and mirroring. Perhaps sanity can be turned into an art form and a science. Belief Revision and Truth Maintenance enable systems to unravel confusion caused by the injection of mutually inconsistent inputs. Nobody's fool, these systems let the user know that there's a fib in there somewhere. Psychology of computers becomes an issue, and the Silicon Syndrome of Neuroses can be detected whenever the machines are not taught how to think straight. Machines are already sapient. Soon they will acquire sentience, and maybe even free will (nothing more than a random number generator coupled with a value system). Perhaps by the end of the Millenium (just 14 years away), the planet will see its first Artificial Sentient Being. Perhaps Von Neumann knew what he was talking about when he wrote his cryptic volume entitled, On the Theory of Self-Reproducing Automata. There were no Cybernauts in Philadelphia this year, but many of the piece parts were in evidence. Perhaps it is just a matter of time until the Golem takes its first step. In the mean time, we have entered the era of the Competent System, somewhat short on world-class expertise, but able to hold it's own in today's corporate culture. It learns about as fast as its human counterpart, and is infinitely clonable. Once upon a time it was felt that machines should work and people should think. Now that machines can think, perhaps people can take more time to enjoy the state of being called Life. * * * Lincroft, NJ August 17, 1986 ------------------------------ End of AIList Digest ******************** From csnet_gateway Sat Sep 20 00:54:25 1986 Date: Sat, 20 Sep 86 00:54:11 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #186 Status: R AIList Digest Friday, 19 Sep 1986 Volume 4 : Issue 186 Today's Topics: Cognitive Science - Commentaries on the State of AI ---------------------------------------------------------------------- Date: 29 Aug 86 01:58:30 GMT From: hsgj@tcgould.tn.cornell.edu (Mr. Barbecue) Subject: Re: Notes on AAAI '86 (not really a followup article, more of a commentary) I find it very interesting that there is so much excitement generated over parallel processing computer systems by the AI community. Interesting in that the problems of AI (the intractability of: language, vision, and general cognition to name a few) are not anywhere near limited by computational power but by our lack of understanding. If somebody managed to create a truely intelligent system, I think we would have heard about it by now, even if the program took a month to run. Fact of the matter is that our knowledge of such problems is minimal. Attempts to solve them leads to researchers banging their heads against a very hard wall, indeed. So what is happening? The field that was once A.I. is very quickly headed back to it's origins in computer science and is producing "Expert Systems" by the droves. The problem isn't that they aren't useful, but rather that they are being touted as the A.I., and true insights into actual human thinking are still rare (if not non-existant). Has everybody given up? I doubt it. However, it seems that economic reality has set in. People are forced to show practical systems with everyday appli- cations. Financers can't understand why we would be overjoyed if we could develop a system that learns like a baby, and so all the money is being siphoned away and into robotics, Expert Systems, and even spelling checkers! (no, I don't think that welding cars together requires a great deal of true intelligence, though technically it may be great) So what is one to do? Go into cog-psych? At least psychologists are working on the fundamental problems that AI started, but many seem to be grasping at straws, trying to find a simple solution (i.e., familly resemblance, primary attribute analysis, etc.) What seems to be lacking is a cogent combination of theories. Some attempts have been made, but these authors basically punt on the issue, stating like "none of the above theories adequately explain the observed phenomena, perhaps the solution is a combination of current hypothesis". Very good, now lets do that research and see if this is true! My opinion? Well, some current work has dealt with computer nervous systems, (Science --sometime this summer). This is similar in form to the hypercube systems but the theory seems different. Really the work is towards computer neurons. Distributed systems in which each element contributes a little to the final result. Signals are not binary, but graded. They combine with other signals from various sources and form an output. Again, this could be done with a linear machine that hold partial results. But, I'm not suggesting that this alone is a solution, it's just interesting. My real opinion is that without "bringing baby up" so to speak, we won't get much accomplished. The ultimate system will have to be able to reach out, grasp (whether visually or physically, or whatever) and sense it's world around it in a rich manner. It will have to be malleable, but still have certain guidelines built in. It must truely learn, forming a myriad of connections with past experiences and thoughts. In sum, it will have to be a living animal (though made of sand..) Yes, I do think that you need the full range of systems to create a truely intelligent system. Hellen Keller still had touch. She could feel vibrations, and she could use this information to create a world that was probably perceptually much different than ours. But, she had true intelligence. (I realize that the semantics of all these words and phrases are highly debated, you know what I'm talking, so don't try to be difficult!) :) Well, that's enough for a day. Ted Inoue. Cornell -- ARPA: hsgj%vax2.ccs.cornell.edu@cu-arpa.cs.cornell.edu UUCP: ihnp4!cornell!batcomputer!hsgj BITNET: hsgj@cornella ------------------------------ Date: 1 Sep 86 10:25:25 GMT From: ulysses!mhuxr!mhuxt!houxm!hounx!kort@ucbvax.Berkeley.EDU (B.KORT) Subject: Re: Notes on AAAI '86 I appreciated Ted Inoue's commentary on the State of AI. I especially agree with his point that a cogent combination of theories is needed. My own betting card favors the theories of Piaget on learning, coupled with the modern animated-graphic mixed-initiative dialogues that merge the Socratic-style dialectic with inexpensive PC's. See for instance the Mind Mirror by Electronic Arts. It's a flashy example of the clever integration of Cognitive Psychology, Mixed Initiative Dialogues, Color Animated Graphics, and the Software/Mindware Exchange. Such illustrations of the imagery in the Mind's Eye can breathe new life into the relationship between silicon systems and their carbon-based friends. Barry Kort hounx!kort ------------------------------ Date: 4 Sep 86 21:39:37 GMT From: tektronix!orca!tekecs!mikes@ucbvax.Berkeley.EDU (Michael Sellers) Subject: transition from AI to Cognitive Science (was: Re: Notes on AAAI '86) > I find it very interesting that there is so much excitement generated over > parallel processing computer systems by the AI community. Interesting in > that the problems of AI (the intractability of: language, vision, and general > cognition to name a few) are not anywhere near limited by computational > power but by our lack of understanding. [...] > The field that was once A.I. is very quickly headed back to > it's origins in computer science and is producing "Expert Systems" by the > droves. The problem isn't that they aren't useful, but rather that they > are being touted as the A.I., and true insights into actual human thinking > are still rare (if not non-existant). Inordinate amounts of hype have long been a problem in AI; the only difference now is that there is actually a small something there (i.e. knowledge based systems), so the hype is rising to truly unbelievable heights. I don't know that AI is returning to its roots in computer science, probably there is just more emphasis on the area(s) where something actually *works* right now. > Has everybody given up? I doubt it. However, it seems that economic reality > has set in. People are forced to show practical systems with everyday appli- > cations. Good points. You should check out the book "The AI Business" by ...rats, it escapes me (possibly Winston or McCarthy?). I think it was published in late '84 or early '85, and makes the same kinds of points that you're making here, talking about the hype, the history, and the current state of the art and the business. > So what is one to do? Go into cog-psych? At least psychologists are working > on the fundamental problems that AI started, but many seem to be grasping at > straws, trying to find a simple solution (i.e., familly resemblance, primary > attribute analysis, etc.) The Grass is Always Greener. I started out going into neurophysiology, then switched to cog psych because the neuro research is still at a lower level than I wanted, and then became disillusioned because all of the psych work being done seemed to be either super low-level or infeasable to test empirically. So, I started looking into computers, longing to get into the world of AI. Luckily, I stopped before I got to the point you are at now, and found something better (no, besides Amway :-)... > What seems to be lacking is a cogent combination of theories. Some attempts > have been made, but these authors basically punt on the issue, stating > like "none of the above theories adequately explain the observed phenomena, > perhaps the solution is a combination of current hypothesis". Very good, now > lets do that research and see if this is true! And this is exactly what is happening in the new field of Cognitive Science. While there is still no "cogent combination of theories", things are beginning to coalesce. (Pylyshyn described the current state of the field as Physics searching for its Newton. Everyone agrees that the field needs a Newton to bring it all together, and everyone thinks that he or she is probably that person. The problem is, no one else agrees with you, except maybe your own grad students.) Cog sci is still emerging as a separate field, even though its beginnings can probably be pegged as being in the late '70s or early '80s. It is taking material, paradigms, and techniques from AI, neurology, cog psych, anthropology, linguistics, and several other fields, and forming a new field dedicated to the study of cognition in general. This does not mean that cognition should be looked at in a vacuum (as is to some degree the case with AI), but that it can and should be examined in both natural and artificial contexts, allowing for the difference between them. It can and should take into account all types and levels of cognition, from the low-level neural processing to the highly plastic levels of linguistic and social cognitive interaction, researching and applying these areas in artificial settings as it becomes feasable. > [...] My real opinion is that > without "bringing baby up" so to speak, we won't get much accomplished. The > ultimate system will have to be able to reach out, grasp (whether visually or > physically, or whatever) and sense it's world around it in a rich manner. It > will have to be malleable, but still have certain guidelines built in. It > must truely learn, forming a myriad of connections with past experiences and > thoughts. In sum, it will have to be a living animal (though made of sand..) This is one possibility, though not the only one. Certainly an artificially cogitating system without many of the abilities you mention would be different from us, in that its primary needs (food, shelter, sensory input) would not be the same. This does not make these things a requirement, however. If we would wish to build an artificial cogitator that had roughly the same sort of world view as we have, then we probably would have to give it some way of directly interacting with its environment through the use of sensors and effectors of some sort. I suggest that you find and peruse the last 5 or 6 years of the journal Cognitive Science, put out by the Cognitive Science Society. Most of the things that have been written in there are still fairly up-to-date, as the field is still reaching "critical mass" in terms of theoretical quantity and quality (an article by Norman, "Twelve Issues for Cognitive Science" from 1980 in this journal (not sure which issue) discusses many of the things you are talking about here). Let's hear more on this subject! > Ted Inoue. > Cornell -- Mike Sellers UUCP: {...your spinal column here...}!tektronix!tekecs!mikes INNING: 1 2 3 4 5 6 7 8 9 TOTAL IDEALISTS 0 0 0 0 0 0 0 0 0 1 REALISTS 1 1 0 4 3 1 2 0 2 0 ------------------------------ Date: 6 Sep 86 19:09:31 GMT From: craig@think.com (Craig Stanfill) Subject: Re: transition from AI to Cognitive Science (was: Re: Notes on AAAI '86) > I find it very interesting that there is so much excitement generated over > parallel processing computer systems by the AI community. Interesting in > that the problems of AI (the intractability of: language, vision, and general > cognition to name a few) are not anywhere near limited by computational > power but by our lack of understanding. [...] For the last year, I have been working on AI on the Connection Machine, which is a massively parallel computer. Depending on the application, the CM is between 100 and 1000 times faster than a Symbolics 36xx. I have performed some experiments on models of reasoning from memory (Memory Based Reasoning, Stannfill and Waltz, TMC Technical Report). Some of these experiments required 5 hours on a 32,000 processor CM. I, for one, do not consider a 500-5000 hour experiment on a Symbolics a practical way to work. More substantially, having a massively parallel machine changes the way you think about writing programs. When certain operations become 1000 times faster, what you put into the inner loop of a program may change drasticly. ------------------------------ Date: 7 Sep 86 16:46:51 GMT From: clyde!watmath!watnot!watdragon!rggoebel@CAIP.RUTGERS.EDU (Randy Goebel LPAIG) Subject: Re: transition from AI to Cognitive Science (was: Re: Notes on AAAI '86) Mike Sellers from Tektronix in Wilsonville, Oregon writes: | Inordinate amounts of hype have long been a problem in AI; the only difference | now is that there is actually a small something there (i.e. knowledge based | systems), so the hype is rising to truly unbelievable heights. I don't know | that AI is returning to its roots in computer science, probably there is just | more emphasis on the area(s) where something actually *works* right now. I would like to remind all that don't know or have forgotten that the notion of a rational artifact as digitial computer does have its roots in computing, but the more general notion of intelligent artifact has concerned scientists and philosophers much longer than the lifetime of the digital computer. John Haugeland's book ``AI: the very idea'' would be good reading for those who aren't aware that there is a pre-Dartmouth history of ``AI.'' Randy Goebel U. of Waterloo ------------------------------ End of AIList Digest ******************** From csnet_gateway Sat Sep 20 00:54:41 1986 Date: Sat, 20 Sep 86 00:54:35 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #187 Status: R AIList Digest Friday, 19 Sep 1986 Volume 4 : Issue 187 Today's Topics: Queries - Natural Language DB Interface & NL Generation & Production Systems & Facial Recognition & Smalltalk & Symbolics CAD & Lisp Machine News & MACSYMA & San Diego Speakers Wanted ---------------------------------------------------------------------- Date: 16 Sep 86 20:05:31 GMT From: mnetor!utzoo!utcs!bnr-vpa!bnr-di!yali@seismo.css.gov Subject: natural language DB interface Has anyone out there any experience with the Swan* natural language database interface put out by Natural Language Products of Berkeley? This system was demo-ed at AAAI this August. I am primarily interested in the system's ability to talk to "different databases associated with different DBMS's" simultaneously (quoting an information sheet put out by NLP). How flexible is it and how easy is it to adapt to new domains? ====================================================== Yawar Ali {the world}!watmath!utcsri!utcs!bnr-vpa!bnr-di!yali ====================================================== * Swan is an unregistered trademark of NLP ------------------------------ Date: Thu, 18 Sep 86 16:34:47 edt From: lb0q@andrew.cmu.edu (Leslie Burkholder) Subject: natural language generation Has work been done on the problem of generating relatively idiomatic English from sentences written in a language for first-order predicate logic? Any pointers would be appreciated. Leslie Burkholder lb0q@andrew.cmu.edu ------------------------------ Date: Thu, 18 Sep 1986 17:10 EDT From: LIN@XX.LCS.MIT.EDU Subject: queries about expert systems Maybe some AI guru out there can help with the following questions: 1. Production systems are the implementation of many expert systems. In what other forms are "expert systems" implemented? [I use the term "expert system" to describe the codification of any process that people use to reason, plan, or make decisions as a set of computer rules, involving a detailed description of the precise thought processes used. If you have a better description, please share it.] 2. A production system is in essence a set of rules that state that "IF X occurs, THEN take action Y." System designers must anticipate the set of "X" that can occur. What if something happens that is not anticipated in the specified set of "X"? I assert that the most common result in such cases is that nothing happens. Am I right, wrong, or off the map? Thanks. Herb Lin ------------------------------ Date: 11 Sep 86 20:42:14 GMT From: ihnp4!islenet!humu!uhmanoa!aloha1!ryan@ucbvax.Berkeley.EDU (ryan) Subject: querying a data base using an inference engine This is a sort of banner letting the rest of the world know that we at the Artificial Intelligence Lab at the University of Hawaii are currently looking at the problem of querying a database using AI techniques. We will be using a natural language front end for querying the database. We will appretiate any information from anyone working on or interested in the same. my address is Paul Ryan ...{dual,vortex,ihnp4}!islenet!aloha1!ryan ...nosvax!humu!islenet!aloha1!ryan ------------------------------ Date: Thu, 18 Sep 86 18:55:43 edt From: philabs!micomvax!peters@seismo.CSS.GOV Subject: Computer Vision We are starting a project related to automatic classification of facial features from photographs. If anyone out there has any info/references related to this area please let me hear from you. email: !philabs!micomvax!peters mail: Peter Srulovicz Philips Information Systems 600 Dr. Philips Blvd St. Laurent Quebec Canada H4M-2S9 ------------------------------ Date: 16 Sep 86 01:22:57 GMT From: whuxcc!lcuxlm!akgua!gatech!gitpyr!krubin@bellcore.com Subject: Smalltalk as an AI research tool? I am currently working on an AI project where we are using Smalltalk-80 as our implementation language. Are there others who have used Smalltalk to do serious AI work? If so, and you can talk about what you have done, please respond. I would be interested in learning how well suited the language is for serious AI work. We have plans to implement an (Intelligent Operator Assistant) using an IBM PC-AT running a version of Digitalk Incorporated's Smalltalk/V. Any comments on this software would also be helpful (especially speed information!). Kenneth S. Rubin (404) 894-4318 Center for Man-Machine Systems Research School of Industrial and Systems Engineering Georgia Institute of Technology Post Office Box 35826 Atlanta, Georgia 30332 Majoring with: School of Information and Computer Science ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!krubin ------------------------------ Date: 14 Sep 86 11:35:00 GMT From: osiris!chandra@uxc.cso.uiuc.edu Subject: Wanted: CAD program for Symbolics CAD software for the Symbolics Machine Hi, I just got a Symbolics lisp machine. I am looking for any Public Domain design/drafting program. Being an architect I'd like to draw stuff on my lisp machine. Hints, pointers, references would be appreciated. Thanks, navin chandra ARPA: dchandra@athena.mit.edu BITNET: ank%cunyvms1 ------------------------------ Date: 18 Sep 86 03:29:53 GMT From: hp-sdd!ncr-sd!milano!dave@hplabs.hp.com Subject: Lisp Machine News? Does anyone have or know of a zwei-based interface to news? (If it exists, 3 to 2 it's called ZNEWS.) Dave Bridgeland -- MCC Software Technology ARPA: dave@mcc.arpa UUCP: ut-sally!im4u!milano!daveb "Things we can be proud of as Americans: * Greatest number of citizens who have actually boarded a UFO * Many newspapers feature "JUMBLE" * Hourly motel rates * Vast majority of Elvis movies made here * Didn't just give up right away during World War II like some countries we could mention * Goatees & Van Dykes thought to be worn only by weenies * Our well-behaved golf professionals * Fabulous babes coast to coast" ------------------------------ Date: 15 Sep 86 16:17:00 GMT From: uiucuxa!lenzini@uxc.cso.uiuc.edu Subject: Wanted: MACSYMA info Hi, I have a friend in the nuclear eng. department who is currently working on a problem in - I can't remember right now but that's not the point - anyway, this problem involves the analytic solution of a rather complex integral (I believe it's called Chen's (sp?) integral). A while back I heard something about a group of programs called MACSYMA that were able to solve integrals that were previously unsolvable. I suggested that he may want to look into the availabiliy of MACSYMA. I would appreciate any information about these programs - what they can and can't do, how they are used, how to purchase (preferably with a university discount) , etc. Thanks in advance, Andy Lenzini University of Illinois. ...pur-ee!uiucdcs!uiucuxa!lenzini ------------------------------ Date: 18 Sep 86 13:58 PDT From: sigart@LOGICON.ARPA Subject: Speakers wanted The San Diego Special Interest Group on Artificial Intelligence (SDSIGART) is looking for speakers for its regular monthly meetings. We are presently looking for individuals who would like to give a presentation on any AI topic during the January to April 1987 time-frame. We typically hold our meetings on the fourth thursday of the month, and provide for a single presentation during the meeting. If you anticipate being in San Diego during that time and would like to give a presentation please contact us via E-mail at sigart\@logicon.arpa. We cannot provide transportation reimbursement for speakers from outside the San Diego area, but we can provide some reimbursement of hotel/meal expenses. Thank You, Bill D'Camp ------------------------------ End of AIList Digest ******************** From csnet_gateway Sun Sep 21 06:58:05 1986 Date: Sun, 21 Sep 86 06:57:56 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #188 Status: RO AIList Digest Saturday, 20 Sep 1986 Volume 4 : Issue 188 Today's Topics: Education - AI Grad Schools, Philosophy - Discussion of AI and Associative Memory, AI Tools - Turbo Prolog ---------------------------------------------------------------------- Date: 12 Sep 86 20:39:56 GMT From: ihnp4!gargoyle!sphinx!bri5@ucbvax.Berkeley.EDU (Eric Brill) Subject: AI Grad Schools A few weeks ago, I posted a request for info on good graduate schools for AI. I got so many requests to forward the replies I got, that I decided to just post a summary to the net. So here it is: Almost everybody seemed to believe that the top 4 schools are MIT, CMU, Stanford and Yale (not necessarily in that order). Other schools which got at least 2 votes for being in the top 10 were Toronto, Illinois(Urbana), UMass(Amherst), Berkeley, UCLA, UCIrvine, UTexas(Austin). Other schools which got one vote for being in the top 10 were URochester, UCSD, Syracuse and Duke. ------------------------------ Date: Tue, 09 Sep 86 12:00:00 GMT+2 From: H29%DHDURZ2.BITNET@WISCVM.WISC.EDU Subject: AI-discussion In the last AI-lists there has been a discussion about the possibilities of intelligent machines. I am going to add some arguments I missed in the discussion. 1. First is to claim, that there are a lot of cognitive functions of man which can be simulated by the computer. But one problem is, that up to now these different functions are not integrated in one machine or superprogram to my kwowledge. 2. There is the phenomenon of intentionality amd motivation in man that finds no direct correspondent phenomenon in the computer. 3. Man's neuronal processing is more analogue than digital in spite of the fact that neurons can only have two states. Man's organisation of memory is rather associative than categorial. [Neurons are not two-state devices! Even if we ignore chemical and physiological memory correlates and the growth and decay of synapses, there are the analog or temporal effects of potential buildup and the fact that neurons often transmit information via firing rates rather than single pulses. Neurons are nonlinear but hardly bistable. -- KIL] Let me elaborate upon these points: Point 1: Konrad Lorenz assumes a phenomenon he called " fulguration" for systems. This means in the end nothing more than: The whole is more than the sum of parts. If you merge all possible functions a computer can do to simulate human abilities, you will get higher functions which transgress the sum of all lower functions. You may once get a function like consciousness or even selfconscious- ness. If you define self as the man's knowledge of himself: his qualities, abilities, his existence. I see no general problem to feed this knowledge to a computer. Real "understanding" of natural language however needs not only lingui- stic competence but also sensory processing and recognition abilities (visual, acoustical). Language normally refers to objects which we first experience by sensory input and then name it. The construct- ivistic theory of human learning of language by Paul Lorenzen und O. Schwemmer (Erlanger Schule) assumes a "demonstration act" (Zeige- handlung) constituting a fundamental element of man (child) learning language. Without this empirical fundament of language you will never leave the hermeneutic circle, which drove former philosphers into despair. Point 2.: One difference between man and computer is that man needs food and computers need electricity and further on the computer doesn't cry when somebody is going to pull his plug. Nevertheless this can be made: A computer,a robot that attacks every- body by weapon, who tries to pull his plug. But who has an interest to construct such a machine? To living organisms made by evolution is given the primary motivation of self-preservation. This is the natural basis of intentionality. Only the implementation of intentionality, motivation, goals and needs can create a machine that deserves the name "intelligent". It is intelligent by the way it reaches "his" goals. Implementation of "meaning" needs the ability of sensory perception and recognition, linguistical competence and understanding, having or simulating intentions. To know the meaning of an object means to understand the function of this object for man in a means-end relation within his living context. It means to realize for which goals or needs the "object" can be used. Point 3.: Analogue information processing may be totally simulated by digitital processing or may be not. Man's associative organization of memory, however needs storage and retrieval mechanism other than those now available or used by computers. I have heard that some scientists try to simulate associative memory organization in the states, but I have no further information about that. (Perhaps somebody can give me information or references. Thanks in advance!). [Geoffrey E. Hinton and James A. Anderson (eds.), Parallel Models of Associative Memory, Lawrence Erlbaum Associates, Inc., Hillsdale NJ. Dr. Hinton is with the Applied Psychology Unit, Cambridge England. -- KIL] Scientists working on AI should have an attitude I call "critical opti- mism". This means being critical,see the problems and not being euphoric, that all problems can be solved in the next ten years. On the other hand it means not to assume any problem as unsolvable but to be optimistic, that the scientific community will solve problems step by step, one after the other how long it will ever last. Finally let me - being a psychologist - state some provocative hypotheses: The belief, that man's cognitive or intelligent abilities including having intentions will never be reached by a machine, is founded in the conscious or unconscious assumption of man's godlike or godmade uniqueness, which is supported by the religious tradition of our culture. It needs a lot of self-reflection, courage and consciousness about one's own existential fears to overcome the need of being unique. I would claim, that the conviction mentioned above however philosphical or sophisticated it may be justified, is only the "RATIONALIZATION" (in the psychoanalytic meaning of the word) of understandable but irrational and normally unconscious existential fears and need of human being. PETER PIRRON MAIL ADDRESS: Psychologisches Institut Hauptstrasse 49-53 D-6900 Heidelberg Western Germany ------------------------------ Date: Thu 18 Sep 86 20:04:31-CDT From: CS.VANSICKLE@R20.UTEXAS.EDU Subject: What's wrong with Turbo Prolog 1. Is Borland's Turbo Prolog a superset of the Clocksin & Mellish (C & M) standard? On the back cover of Turbo Prolog's manual is the description "A complete Prolog incremental compiler supporting a large superset of Clocksin & Mellish Edinburgh standard Prolog." This statement is not true. On page 127 the manual says "Turbo Prolog . . . contains virtually all the features described in Programming in Prolog by Clocksin and Mellish." If you read "virtually" as "about 20% of" then this statement is true. Turbo Prolog does use Edinburgh syntax, that is, :- for "if" in rules, capitalized names for variables, lower case names for symbols, square brackets for delimiting lists, and | between the head and tail of a list. Almost all the Clocksin & Mellish predicates have different names, different arguments, or are missing entirely from Turbo Prolog. For example, "var" is "free," and "get0" is "readchar." Differences in predicate names and arguments are tolerable, and can be handled by a simple conversion program or by substitutions using an editor. They could also be handled by adding rules that define the C & M predicates in terms of Turbo Prolog predicates, for example, var(X):-free(X). These kinds of differences are acceptable in different implementations of Prolog. Even C & M say that their definition should be considered a core set of features, that each implementation may have different syntax. However, Borland has done much more than just rename a few predicates. 2. Is Borland's Turbo Prolog really Prolog? NO. Turbo Prolog lacks features that are an essential part of any Prolog implementation and requires declarations. Borland has redefined Prolog to suit themselves, and not for the better. A key feature of Lisp and Prolog is the ability to treat programs and data identically. In Prolog "clause," "call," and "=.." are the predicates that allow programs to be treated as data, and these are missing entirely from Turbo Prolog. One use of this feature is in providing "how" and "why" explanations in an expert system. A second use is writing a Prolog interpreter in Prolog. This is not just a theoretically elegant feature, it has practical value. For a specific domain a specialized interpreter can apply domain knowledge to speed up execution, or an intelligent backtracking algorithm could be implemented. In C & M Prolog a Prolog interpreter takes four clauses. Borland gives an example "interpreter" on page 150 of the Turbo Prolog manual - nine clauses and twenty-two declarations. However, their "interpreter" can't deal with any clause, it can only deal with "clauses" in a very special form. A clause such as likes(ellen,tennis) would have to be represented as clause(atom(likes,[symbol(ellen),symbol(tennis)]),[]) in Borland's "interpreter." I don't expect "clause" to retrieve compiled clauses, but I do expect Prolog to include it. By dropping it Borland has introduced a distinction between programs and data that eliminates a key feature of Prolog. Turbo Prolog absolutely requires data typing. Optional typing would be a good feature for Prolog - it can produce better compiled code and help with documentation. However, required typing is not part of any other Prolog implementation that I know of. Typing makes life easier for the Turbo Prolog compiler writer at the expense of the Turbo Prolog programmers. A little more effort by the few compiler writers would have simplified the work of the thousands of potential users. There are good Prolog compilers in existence that do not require typing, for example, the compiler for DEC-10 Prolog. It may also be that Borland thought they were improving Prolog by requiring typing, but again, why not make it optional? Besides introducing a distinction between programs and data, Turbo Prolog weakens the ability to construct terms at run time. One of the great strengths of Prolog is its ability to do symbolic computation, and Borland has seriously weakened this ability. Again this omission seems to be for the convenience of the compiler writers. There are no predicates corresponding to the following C & M predicates, even under other names: "arg," "functor," "name," "=..," "atom," "integer," and "atomic." These predicates are used in programs that parse, build, and rewrite structured terms, for example, symbolic integration and differentiation programs, or a program that converts logical expressions to conjunctive normal form. The predicate "op" is not included in Turbo Prolog. Full functional notation must be used. You can write predicates to pretty print terms, and the manual gives an example of this, but it is work that shouldn't be necessary. Dropping "op" removed one of Prolog's strongest features for doing symbolic computation. Turbo Prolog introduces another distinction between clauses defined at compile time and facts asserted at run time. Apparently only ground terms can be asserted, and rules cannot be asserted. This may be partly a result of having only a compiler and no interpreter. The predicates for any facts to be asserted must be declared at compile time. This is another unecessary distinction for the convenience of the compiler writers. One other annoyance is the lack of DCG rules, and the general difficulty of writing front ends that translate DCG rules and other "syntactic sugar" notations to Prolog rules. 3. Is Turbo Prolog suitable for real applications? I think Turbo Prolog could run some real applications, but one limitation is that a maximum of 500 clauses is allowed for each predicate. One real application program computes the intervals of a directed graph representing a program flow graph. Each node represents a program statement, and each arc represents a potential transfer of control from the head node to the tail node. There is a Prolog clause for each node and a clause for each arc. A program with 501 statements would exceed Turbo Prolog's limit. I assume Borland could increase this limit, but as it stands, this is one real application that Turbo Prolog would not run. 4. Is there anything good about Turbo Prolog? YES. I like having predicates for windows, drawing, and sound. It looks easy to write some nice user interfaces using Turbo Prolog's built in predicates. The manual is well done, with lots of examples. There is easy access to the facilities of MS-DOS. There is a good program development environment, with windows for editing code, running the program, and tracing. There are also features for allowing programming teams to create applications - modules and separate name spaces. At $100 the price is right. If this were Prolog, it would be a great product. -- Larry Van Sickle cs.vansickle@r20.utexas.edu 512-471-9589 ------------------------------ Date: Thu 18 Sep 86 20:12:29-CDT From: CS.VANSICKLE@R20.UTEXAS.EDU Subject: Simple jobs Turbo Prolog can't do Two simple things you CANNOT do in Turbo Prolog: 1. Compute lists containing elements of different basic types. Turbo Prolog does not let you have goals such as append([a,b,c],[1,2,3],L). Turbo Prolog requires that the types of every predicate be declared, but the typing system does not allow you to declare types that mix basic types. Also lists like: [1,a] [2,[3,4]] [5,a(6)] cannot be created in Turbo Prolog. The syntax of types is: a) name = basictype where basictype is integer, char, real, string or symbol, b) name = type* where type is either a basic type or a user defined type, the asterisk indicates a list, c) name = f1(d11,...d1n1);f2(d21,...,d2n2);...fm(dm1,...d2nm) where fi are functors and dij are types, called "domains." The functors and their domains are alternative structures allowed in the type being defined. The important thing to notice is that you cannot define a type that has basic types as alternatives. You can only define alternatives for types that contain functors. So you cannot define types mytype = integer;symbol mylisttype = mytype* which is what you would need to append a list of integers to a list of symbols. What the Turbo Prolog manual recommends for this case is to define mytype = s(symbol);i(integer) mylisttype = mytype* and declare append as append(mylisttype,mylisttype,mylisttype) which would allow us to state the goal append([s(a),s(b),s(c)],[i(1),i(2),i(3)],L). This is clumsy, kludgy, and ugly. 2. Compute expressions that contain different basic types or mixtures of structures and basic types. Simplifying arithmetic expressions that contain constants and variables seems like it should be easy in a language designed to do symbolic computation. In C & M Prolog some rules for simplifying multiplication might be simplify(0 * X,0). simplify(X * 0,0). simplify(1 * X,X). simplify(X * 1,X). In C & M Prolog you can enter goals such as simplify(a - 1 * (b - c),X). Now in Turbo Prolog, because of the limited typing, you cannot have expressions that contain both symbols and integers. (You also cannot have infix expressions, but that is another issue). Instead, you would have to do something like this: exprtype = i(integer);s(symbol);times(exprtype,exprtype) and declare simplify as: simplify(exprtype,exprtype) and the clauses would be: simplify(times(i(0),X),i(0)). simplify(times(X,i(0)),i(0)). simplify(times(i(1),X),X). simplify(times(X,i(1)),X). The goal would be: simplify(minus(s(a),times(i(1),minus(s(b),s(c)))),X). This should speak for itself, but I'll spell it out: REAL Prolog can do symbolic computation involving mixtures of symbols, numeric constants, and expressions; the programs are simple and elegant; input and output are easy. In Turbo Prolog you can't even create most of the expressions that real Prolog can; the programs are long, opaque, and clumsy; you have to write your own predicates to read and write expressions in infix notation. It is a shame that this product comes from a company with a reputation for good software. If it came from an unknown company people would be a lot more cautious about buying it. Since it's from Borland, a lot of people will assume it's good. They are going to be disappointed. -- Larry Van Sickle cs.vansickle@r20.utexas.edu 512-471-9589 ------------------------------ End of AIList Digest ******************** From csnet_gateway Sun Sep 21 06:57:50 1986 Date: Sun, 21 Sep 86 06:57:42 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #189 Status: RO AIList Digest Saturday, 20 Sep 1986 Volume 4 : Issue 189 Today's Topics: AI Tools - Symbolics Lisp Release 7, Games - Connect Four & Computer Chess News, Applications - Music-Research Digest, Contest - New Deadline for HP Contest ---------------------------------------------------------------------- Date: 2 Sep 86 20:03:34 GMT From: mcvax!euroies!ceri@seismo.css.gov (Ceri John Fisher) Subject: Symbolics Lisp Release 7 (Genera) Information requested: Does anybody have any concrete comments on Symbolics Release 7 Common Lisp and ZetaLisp and new window system. We have release 6 and are rather fear- fully awaiting the next release since we have started to hear rumours of large resources required and relatively poor performance (in spite of increased ease of use). Can anyone confirm or deny this from actual experience ? Mail me with your comments and I will summarize to the net if there's enough interest. Thank you for your attention. Ceri Fisher, Plessey (UK) Ltd, Christchurch, England. ceri@euroies.UUCP or ..!mcvax!ukc!euroies!ceri -- currently under revision ------------------------------ Date: 6 Sep 86 19:02:10 GMT From: well!jjacobs@hplabs.hp.com (Jeffrey Jacobs) Subject: Re: Symbolics Lisp Release 7 (Genera) You want Common Lisp, you gotta pay the price ! I've heard the same rumors... ------------------------------ Date: 15 Aug 86 16:11:25 GMT From: mcvax!botter!klipper!victor@seismo.css.gov (L. Victor Allis) Subject: Information wanted. I'm looking for any information I can get on a game which is a more complex kind of tic-tac-toe. In the Netherlands this game is called 'vier op een rij', in Germany 'vier gewinnt'. [Here it's marketed by Milton Bradley as Connect Four. -- KIL] Description of the game: 'Vier op een rij' is played on a vertical 6 x 7 grid. Two players, white and black, the former having 21 white, the latter having 21 black stones, play the game by alternately throwing one of their stones in one of the 7 vertical columns. The stone will fall down as far as possible. The goal of the game is to have four of your stones on four consecutive horizontal, vertical or diagonal positions (like tic-tac-toe). The one who achieves this first, wins. A draw is possible, if none achieved this and the grid is full. White always has the first 'move'. It is not allowed to pass. Possible situation in a game: --------------- | | | | | | | | White (x) will lose this game since in this --------------- situation he has to play the second column to | | | |o| | | | prevent black (o) from winning horizontaly, --------------- but this will give black the possibility to | | | |x| | | | win diagonaly by playing the second column again. --------------- | | |o|o|o| | | --------------- | |x|x|o|x| | | --------------- |o|x|x|x|o| | | --------------- I would like to know if there is someone who wrote a program for this game and any results which were obtained by this program, like: 1) Result of the game after perfect play of both sides. 2) Best opening moves for both sides. Thanks ! Victor Allis. victor@klipper.UUCP Free University of Amsterdam. The Netherlands. ------------------------------ Date: 18 Aug 86 23:27:15 GMT From: ihnp4!cuae2!ltuxa!ttrdc!levy@ucbvax.Berkeley.EDU (Daniel R. Levy) Subject: Re: Information wanted. In article <585@klipper.UUCP>, victor@klipper.UUCP (L. Victor Allis) writes: >I'm looking for any information I can get on a game which is a >more complex kind of tic-tac-toe. In the Netherlands this game >is called 'vier op een rij', in Germany 'vier gewinnt'. On this vanilla System V R2 3B20 the game is available as /usr/games/connect4 (sorry, no source code came with it on this UNIX-source-licensed system and even if it did it might be proprietary [ :-) ] but I hope this pointer is better than nothing). Please excuse me for posting rather than mailing. My route to overseas sites seems tenuous at best. ------------------------------- Disclaimer: The views contained herein are | dan levy | yvel nad | my own and are not at all those of my em- | an engihacker @ | ployer or the administrator of any computer | at&t computer systems division | upon which I may hack. | skokie, illinois | -------------------------------- Path: ..!{akgua,homxb,ihnp4,ltuxa,mvuxa, go for it! allegra,ulysses,vax135}!ttrdc!levy ------------------------------ Date: 4 Sep 86 21:42:59 GMT From: ubc-vision!alberta!tony@uw-beaver.arpa (Tony Marsland) Subject: Computer Chess News The June 1986 issue of the ICCA Journal is now being distributed. The issue contains the following articles: "Intuition in Chess" by A.D. de Groot "Selective Search without Tears" by D. Beal "When will Brute-force Programs beat Kasparov?" by D. Levy Also there is a complete report on the 5th World Computer Chess Championship by Helmut Horacek and Ken Thompson, including all the games. There are many other short articles, reviews and news items. Subscriptions available from: Jonathan Schaeffer, Computing Science Dept., Univ. of Alberta, Edmonton T6G 2H1, Canada. Cost: $15 for all four 1985 issues $20 per year beginning 1987, $US money order or check/cheque. email: jonathan@alberta.uucp for more information. ------------------------------ Date: Sat, 30 Aug 86 11:00:56 GMT From: Stephen Page Subject: New list: Music-Research Digest COMPUTERS AND MUSIC RESEARCH An electronic mail discussion group The Music-Research electronic mail redistribution list was established after a suggestion made at a meeting in Oxford in July 1986, to provide an effective and fast means of bringing together musicologists, music analysts, computer scientists, and others working on applications of computers in music research. Initially, the list was established for people whose chief interests concern computers and their applications to - music representation systems - information retrieval systems for musical scores - music printing - music analysis - musicology and ethnomusicology - tertiary music education - databases of musical information The following areas are not the principal concern of this list, although overlapping subjects may well be interesting: - primary and secondary education - sound generation techniques - composition There are two addresses being used for this list: - music-research-request@uk.ac.oxford.prg for requests to be added to or deleted from the list, and other administrivia for the moderator. - music-research@uk.ac.oxford.prg for contributions to the list. The above addresses are given in UK (NRS) form. For overseas users, the INTERNET domain-style name for the moderator is music-research-request@prg.oxford.ac.uk If your mailer does not support domain-style addressing, get it fixed. For the moment, explicitly send via the London gateway, using music-research-request%prg.oxford.ac.uk@cs.ucl.ac.uk or music-research-request%prg.oxford.ac.uk@ucl-cs.arpa UUCP users who do not have domain-style addressing may send via Kent: ...!ukc!ox-prg!music-research-request ------------------------------ Date: 8 Sep 86 19:46:47 GMT From: hpcea!hpfcdc!hpfclp!hpai@hplabs.hp.com (AI) Subject: New deadline for HP contest [Forwarded from the Prolog Digest by Laws@SRI-STRIPE.] Hewlett-Packard has extended the submittal deadline for its AI programming contest. Software and entry forms must by sent on or before February 1, 1987. In addition, originality has been added as a judging criterion. That is, newly written software will be weighted more heavily than ported software. Revised rules and an entry form follow. Hewlett-Packard AI Programming Contest To celebrate the release of its AI workstation, Hewlett-Packard is sponsoring a programming contest. Submit your public domain software by February 1, 1987 to be considered for the following prizes: First prize: One HP72445A computer (Vectra) Second prize: One HP45711B computer (Portable Plus) Third prize: One HP16C calculator (Computer Scientist) Complete rules follow. 1. All entries must be programs of interest to the symbolic computing or artificial intelligence communities. They must be executable on HP9000 Series 300 computers running the HP-UX operating system. This includes programs written in the Common LISP, C, Pascal, FORTRAN, or shell script languages, or in any of our third party AI software. 2. All entries must include source code, machine-readable documentation, a test suite, and any special instructions necessary to run the software. Entries may be submitted by electronic mail or shipped on HP formatted 1/4" Certified Data Cartridge tapes. 3. All entries must be in the public domain and must be accompanied by an entry form signed by the contributor(s). Entries must be sent on or before February 1, 1987. 4. Only residents of the U.S. may enter. HP employees and their dependents are ineligible to receive prizes, but are welcome to submit software. In the case of team entries, each member of the team must be eligible. No duplicate prizes will be awarded. Disposition of the prize is solely the responsibility of the winning team. 5. Entries will be judged on the basis of originality, relevance to our user community, complexity, completeness, and ease of use. The date of receipt will be used as a tie-breaker. Decision of the judges will be final. 6. HP cannot return tape cartridges. 7. Selected entries will be distributed by HP on an unsupported software tape. This tape will be available from HP for a distribution fee. The contributor(s) of each entry which is selected for this tape will receive a complimentary copy. To enter: Print and complete the following entry form and mail it to: AI Programming Contest M.S. 99 Hewlett-Packard 3404 E. Harmony Road Fort Collins, CO 80525 Send your software on HP formatted 1/4"tape to the same address, or send it via electronic mail to: hplabs!hpfcla!aicontest or ihnp4!hpfcla!aicontest [Form deleted: write to the author or check the Prolog Digest. I generally omit entry forms and conference reservation coupons to save bandwidth, reduce storage space, and avoid annoying those with slow terminals or expensive communication links. -- KIL] ------------------------------ End of AIList Digest ******************** From csnet_gateway Sat Sep 20 18:52:44 1986 Date: Sat, 20 Sep 86 18:52:34 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #190 Status: RO AIList Digest Saturday, 20 Sep 1986 Volume 4 : Issue 190 Today's Topics: AI Tools - Xerox Dandelion vs. Symbolics ---------------------------------------------------------------------- Date: 4 Sep 86 14:27:00 GMT From: princeton!siemens!steve@CAIP.RUTGERS.EDU Subject: Xerox Dandelion vs. Symbolics? Why do people choose Symbolics/ZetaLisp/CommonLisp over Xerox Dandelion/Interlisp? I have been "brought up" on Interlisp and had virtually no exposure to Maclisp derivatives, but more to the point, I've been brought up on the Xerox Dandelion lisp machine and never used a Symbolics. Every chance I get, I try to find out what a Symbolics/Zetalisp machine has that the Dandelion doesn't. So far I have found only the following: 1) More powerful machine (but less power per dollar). 2) The standard of Commonlisp (only the past couple years). 3) People are ignorant of what the Dandelion has to offer. 4) Edit/debug cycle (and editor) very similar to old standard systems such as Unix/C/Emacs or TOPS/Pascal/Emacs, and therefore easier for beginners with previous experience. I have found a large number of what seem to be advantages of the Xerox Dandelion Interlisp system over the Symbolics. I won't post anything now because this already is too much like an ad for Xerox, but you might get me to post some separately. I am not personally affiliated with Xerox (although other parts of my company are). I am posting this because I am genuinely curious to find out what I am missing, if anything. By the way, the Interlisp system on the Dandelion is about 5 megabytes (it varies depending on how much extra stuff you load in - I've never seen the system get as large as 6 Mb). I hear that Zetalisp is 24 Mb. Is that true? What is in it, that takes so much space? Steven J. Clark, Siemens Research and Technology Laboratory etc. {ihnp4!princeton | topaz}!siemens!steve something like this ought to work from ARPANET: steve@siemens@spice.cs.cmu (i.e. some machines at CMU know siemens). ------------------------------ Date: 5 Sep 86 16:38:57 GMT From: tektronix!orca!tekecs!mikes@ucbvax.Berkeley.EDU (Michael Sellers) Subject: Re: Xerox Dandelion vs. Symbolics? [vs. Tek 4400 series] > Why do people choose Symbolics/ZetaLisp/CommonLisp over > Xerox Dandelion/Interlisp? Maybe I'm getting in over my head (and this is not unbiased), but what about Tek's 4400 series (I think they have CommonLisp & Franz Lisp, but I could be wrong)? I was under the impression that they offered much more bang for the buck than did the other major AI workstation folks. Have you seen these and decided they are not what you want, or are you unaware of their capabilities/cost? > ...Dandelion Interlisp system over the Symbolics. I won't post anything > now because this already is too much like an ad for Xerox, but you might > get me to post some separately. Maybe, if we're going to have testimonials, we could nudge someone from Tek's 4400 group (I know some of them are on the net) into giving us a rundown on their capabilities. > I am not personally affiliated with Xerox (although other parts of my > company are). I am posting this because I am genuinely curious to find > out what I am missing, if anything. I am personally affiliated with Tek (in a paycheck sort of relationship), though not with the group that makes the 4400 series of AI machines. I did have one on my desk for a while, though (sigh), and was impressed. I think you're missing a fair amount :-). > Steven J. Clark, Siemens Research and Technology Laboratory etc. Mike Sellers UUCP: {...your spinal column here...}!tektronix!tekecs!mikes INNING: 1 2 3 4 5 6 7 8 9 TOTAL IDEALISTS 0 0 0 0 0 0 0 0 0 1 REALISTS 1 1 0 4 3 1 2 0 2 0 ------------------------------ Date: 5 Sep 86 17:27:54 GMT From: gatech!royt@seismo.css.gov (Roy M Turner) Subject: Re: Xerox Dandelion vs. Symbolics? In article <25800003@siemens.UUCP> steve@siemens.UUCP writes: > >Every chance I >get, I try to find out what a Symbolics/Zetalisp machine has that the >Dandelion doesn't. So far I have found only the following: >... >Steven J. Clark, Siemens Research and Technology Laboratory etc. >{ihnp4!princeton | topaz}!siemens!steve > As a user of Symbolics Lisp machines, I will try to answer some of Steve's comments. We have had Symbolics machines here since before I started on my degree two years ago; we recently were given thirteen Dandelions and two DandyTigers by Xerox. We use the Symbolics as our research machines, and the Xerox machines for teaching AI. The Symbolics are more powerful, as Steve says, and quite possibly he is right about the power per dollar being less for them than for Xerox; since the Xerox machines were free to us, certainly he's right in our case! :-) However, I find the Dandelions abysmally slow for even small Lisp programs, on the order of the ones we use in teaching (GPS (baby version), micro versions of SAM, ELI, etc.). To contemplate using them for the very large programs that we develop as our research would be absurd--in my opinion, of course. The "standard" of CommonLisp will (so Xerox tells us) be available for the Dandelions soon...'course, they've been saying that for some time now :-). So the two machines may potentially be equal on that basis. ZetaLisp is quite close to CommonLisp (since it was one of the dialects Common Lisp is based on), and also close to other major dialects of lisp--Maclisp, etc.--enough so that I've never had any trouble switching between it and other lisps...with one exception--you guessed it, Interlisp-D. I realize that whatever you are used to colors your thinking, but Lord, that lisp seems weird to me! I mean, comments that return values?? Gimme a break! "People are ignorant of what the Dandelion has to offer." I agree. I'm one of the people. It has nice windows, much less complicated than Symbolics. MasterScope is nice, too. So is the structure editor, but that is not too much of a problem to write on any other lisp machine, and is somewhat confusing to learn (at least, that's the attitude I perceive in the students). What the Dandelions *lack*, however, is any decent file manipulation facilities (perhaps Common Lisp will fix this), a nice way of handling processes, a communications package that works (IP-TCP, at least the copy we received, will trash the hard disk when our UNIX machines write to the DandyTigers...the only thing that works even marginally well is when we send files from the Symbolics! Also, the translation portion of the communication package leaves extraneous line-feeds, etc., lying about in the received file), and A DECENT EDITOR! Which brings us to the next point made by Steve: >4) Edit/debug cycle (and editor) very similar to old standard systems > such as Unix/C/Emacs or TOPS/Pascal/Emacs, and therefore easier > for beginners with previous experience. This is true. However, it is also easier for experts and semi-experts (like me) who may or may not have had prior experience with EMACS. The Dandelions offer a structure editor (and Tedit for text, but that doesn't count) and that's it...if you want to edit something, you do it function by function. Typically, what I do and what other people do on the Xerox machines is enter a function in the lisp window, which makes it very difficult to keep track of what you are doing in the function, and makes it mandatory that you enter one function at a time. Also, the function is immediately evaluated (the defineq is, that is) and becomes part of your environment. Heaven help you if you didn't really mean to do it! At least with ZMACS you can look over a file before evaluating it. Another gripe. Many of our programs used property lists, laboriously entered via the lisp interactor. We do a makefile, and voila--next time we load the file, the properties aren't there! This has yet to happen when something is put in an edit buffer and saved to disk on the Symbolics. Perhaps there is a way of editing on the Xerox machines that lends itself to editing files (and multiple files at once), so that large programs can be entered, edited, and documented (Interlisp-D comments are rather bad for actually documenting code) easily...if so, I haven't found it. Another point in Symbolics favor: reliability. Granted, it sometimes isn't that great for Symbolics, either, but we have had numerous, *numerous* software and hardware failures on the Dandelions. It's so bad that we have to make sure the students save their functions to disk often, and have even had to teach them how to copy sysouts and handle dead machines, since the machines lock up from time to time with no apparent cause. And the students must be cautioned not to save their stuff only to one place, but to save it to the file server, a floppy, and anywhere else they can, since floppies are trashed quite often. Dribble to the hard disk, forget to turn dribble off, there goes the hard disk... Type (logout t) on the Dandelions to cause it not to save your world, and there goes the Dandelion (it works on the DandyTigers). About worlds and sysouts. The Symbolics has a 24-30 meg world, something like that. This is *not* just lisp--it is your virtual memory, just as it is in a Xerox Sysout. The difference in size reflects the amount of space you have at your disposal when creating conses, not the relative sizes of system software (though I imagine ZetaLisp is larger than Interlisp-D). You do not necessarily save a world each time you logout from a Symbolics; you do on a Dandelion...thus the next user who reboots a Symbolics gets a clean lisp, whereas the next user of a Dandelion gets what was there before unless he first copies another sysout and boots off of it. It is, however, much harder to save a world on the Symbolics than on the Xerox machines. Well, I suppose I have sounded like a salesman for Symbolics. I do not mean to imply that Symbolics machines are without faults, nor do I mean to say that Xerox machines are without merit! We are quite grateful for the gift of the Xerox machines; they are useful for teaching. I just tried to present the opinions of one Symbolics-jaded lisp machine user. Back to the Symbolics machine now...I suppose that the DandyTiger beside it will bite me! :-) Roy ------------------------------ Date: 6 Sep 86 22:36:43 GMT From: jade!violet.berkeley.edu!mkent@ucbvax.Berkeley.EDU Subject: Re: Xerox Dandelion vs. Symbolics? As a long-term user of Interlisp-D, I'd be very interested in hearing an *informed* comparison of it with ZetaLisp. However, I'm not particularly interested in hearing what an experienced Zetalisp user with a couple of hours of Interlisp experience has to say on the topic, other than in regard to issues of transfer and learnability. I spent about 4 days using the Symbolics, and my initial reaction was that the user interface was out of the stone age. But I realize this has more to do with *my* background then with Zetalisp itself. Is there anyone out there with *non-trivial* experience with *both* environments who can shed some light on the subject? Marty Kent "You must perfect the Napoleon before they finish Beef Wellington! The future of Europe hangs in the balance..." ------------------------------ Date: 9 Sep 86 06:14:00 GMT From: uiucdcsp!hogge@a.cs.uiuc.edu Subject: Re: Xerox Dandelion vs. Symbolics? >...I spent about 4 days using >the Symbolics, and my initial reaction was that the user interface was out >of the stone age. But I realize this has more to do with *my* background >then with Zetalisp itself. Four days *might* be enough time to get familiarize yourself with the help mechanisms, if that's specifically what you were concentrating on doing. Once you learn the help mechanisms (which aren't bundled all that nicely and are rarely visible on the screen), your opinion of the user interface will grow monotonically with use. If you are interested in having more visible help mechanisms for first-time users, check out what the TI Explorer adds to the traditional Zetalisp environment. LMI and Sperry also provide their own versions of the environment. --John ------------------------------ Date: 10 Sep 86 10:35:40 GMT From: mob@MEDIA-LAB.MIT.EDU (Mario O. Bourgoin) Subject: Re: Xerox Dandelion vs. Symbolics? In article <3500016@uiucdcsp>, hogge@uiucdcsp.CS.UIUC.EDU writes: > >...I spent about 4 days using > >the Symbolics, and my initial reaction was that the user interface was out > >of the stone age..... > > Four days *might* be enough time to get familiarize yourself with the help > mechanisms, if that's specifically what you were concentrating on doing. Four days to learn the help mechanisms? Come on, an acceptable user interface should give you control of help within minutes _not days_. Seriously folks, it took me less than 10 seconds to learn about ZMACS's apropos on the old CADRs and before the end of the day, I knew about a lot more. Have you ever used the "help" key? The Symbolics's software isn't much different from the CADR's. I'll grant that the lispm's presentation of information isn't that obvious or elegant but it isn't stone age and doesn't require 4 days to get a handle on. If you're arguing internals, I haven't worked with the Dandelion so I can't provide an opinion on it. The CADR's user interface software was certainly featureful and appeared to my eyes to come from a different school than what I later saw of Xerox's software. It is useful and manipulable but didn't look intended to be programmed by anyone just off the street. If you want to learn the internals of the user interface, _then_ i'll grant you four days (and more). --Mario O. Bourgoin ------------------------------ Date: 10 Sep 86 15:23:29 GMT From: milano!Dave@im4u.utexas.edu Subject: Re: 36xx vs. Xerox A few to add to pro-36xx list: 5. Reliable hardware 6. Reliable software 7. Good service A year ago, I was on project which used Dandeanimals. As a group, they were up about 60% of the time, and there were days when all 5 were down. The extra screw was that the first level of repair was a photocopier repairman. It always took several days before we got people who knew something about the machines. Dave Bridgeland -- MCC Software Technology (Standard Disclaimer) ARPA: dave@mcc.arpa UUCP: ut-sally!im4u!milano!daveb "Things we can be proud of as Americans: * Greatest number of citizens who have actually boarded a UFO * Many newspapers feature "JUMBLE" * Hourly motel rates * Vast majority of Elvis movies made here * Didn't just give up right away during World War II like some countries we could mention * Goatees & Van Dykes thought to be worn only by weenies * Our well-behaved golf professionals * Fabulous babes coast to coast" ------------------------------ End of AIList Digest ******************** From csnet_gateway Sun Sep 21 06:58:19 1986 Date: Sun, 21 Sep 86 06:58:09 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #191 Status: RO AIList Digest Saturday, 20 Sep 1986 Volume 4 : Issue 191 Today's Topics: AI Tools - Xerox Dandelion vs. Symbolics ---------------------------------------------------------------------- Date: 8 Sep 86 17:35:52 GMT From: hpcea!hpfcdc!hpcnoe!jad@hplabs.hp.com (John Dilley) Subject: Re: Xerox Dandelion vs. Symbolics? [vs. Tek 4400 series] > Why do people choose Symbolics/ZetaLisp/CommonLisp over > Xerox Dandelion/Interlisp? > ... > 3) People are ignorant of what the Dandelion has to offer. I have a file of quotes, one of which has to do with this problem Xerox seems to have. I've heard great things about Dandelion/Interlisp, and their Smalltalk environments, but have never seen one of these machines in "real life" (whatever that is). Anyway, the quote I was referring to is: "It doesn't matter how great the computer is if nobody buys it. Xerox proved that." -- Chris Espinosa And while we're at it ... remember Apple? "One of the things we really learned with Lisa and from looking at what Xerox has done at PARC was that we could construct elegant, simple systems based on just a bit map..." -- Steve Jobs Seems like Xerox needed more advertising or something. It's a shame to see such nice machines go unnoticed by the general public, especially considering what choices we're often left with. -- jad -- John A Dilley Phone: (303)229-2787 Email: {ihnp4,hplabs} !hpfcla!jad (ARPA): hpcnoe!jad@hplabs.ARPA Disclaimer: My employer has no clue that I'm going to send this. ------------------------------ Date: 11 Sep 86 17:58:23 GMT From: gatech!royt@seismo.css.gov (Roy M Turner) Subject: Re: Xerox Dandelion vs. Symbolics? In response to a prior posting by me, Marty (mkent@violet.berkely.edu) writes: > > As a long-term user of Interlisp-D, I'd be very interested in hearing an >*informed* comparison of it with ZetaLisp. However, I'm not particularly >interested in hearing what an experienced Zetalisp user with a couple of >hours of Interlisp experience has to say on the topic... > ... Who, me? :-) If I was unclear in my posting, I apologize. I have had a bit more than two hours of experience w/ Dandelions. I used them in a class I was taking, and also was partly responsible for helping new users and for maintaining some of the software on them. Altogether about 4 months of fairly constant use. Another posting said we were using outdated software; that is undoubtedly correct, as we just got Coda; we were using Intermezzo. Some problems are probably fixed. However, we have not received the new ip-tcp from Xerox...but, what do you expect with free machines? :-) Roy Above opionions my own...'course, they *should* be everyone's! :-) Roy Turner School of Information and Computer Science Georgia Insitute of Technology, Atlanta Georgia, 30332 ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!royt ------------------------------ Date: 12 Sep 86 14:58:07 GMT From: wdmcu%cunyvm.bitnet@ucbvax.Berkeley.EDU Subject: Re: Xerox Dandelion vs. Symbolics? In article <3500016@uiucdcsp>, hogge@uiucdcsp.CS.UIUC.EDU says: >Once you learn the help mechanisms (which aren't bundled all that nicely and >are rarely visible on the screen), your opinion of the user interface will >grow monotonically with use. If you are interested in having more visible ^^^^^^^^^^^^^ Could you please define this word in this context. Thanks. (This is a serious question) /*--------------------------------------------------------------------*/ /* Bill Michtom - work: (212) 903-3685 home: (718) 788-5946 */ /* */ /* WDMCU@CUNYVM (Bitnet) Timelessness is transient */ /* BILL@BITNIC (Bitnet) */ /* */ /* Never blame on malice that which can be adequately */ /* explained by stupidity. */ /* A conclusion is the place where you got tired of thinking. */ /*--------------------------------------------------------------------*/ ------------------------------ Date: 12 Sep 86 07:31:00 GMT From: uiucdcsp!hogge@a.cs.uiuc.edu Subject: Re: Xerox Dandelion vs. Symbolics? >> Four days *might* be enough time to get familiarize yourself with the help >> mechanisms, if that's specifically what you were concentrating on doing. > >Four days to learn the help mechanisms? Come on, an acceptable user >interface should give you control of help within minutes _not days_. >Seriously folks, it took me less than 10 seconds to learn about >ZMACS's apropos on the old CADRs and before the end of the day, I knew >about a lot more. Have you ever used the "help" key? >software isn't much different from the CADR's. I'll grant that the >lispm's presentation of information isn't that obvious or elegant but >it isn't stone age and doesn't require 4 days to get a handle on. There's more subtle help available on the machine than just the help key, and my experience is that it takes a long time for one to learn the mechanisms that are there. The HELP key *is* the main source of help, but not the only source. Examples include: 1. use of Zmacs meta-point to find examples of how to do things (such as hack windows) from the system source, 2. use of c-/ in the Zmacs minibuffer for listing command completions (and what a drag if you don't know about this command) 3. the importance of reading who-line documentation 4. use of the Apropos function to hunt down useful functions, as well as WHO-CALLS 5. use of the various Lisp Machine manufacturer's custom help mechanisms, such as the Symbolics flavor examiner and documentation examiner, or TI's Lisp-completion input editor commands and Suggestions Menus. The Lisp Machine is a big system, and there's lots of good help available. But it isn't trivial learning how to get it nor when to seek it. --John ------------------------------ Date: 12 Sep 86 14:42:58 GMT From: ihnp4!wucs!sbc@ucbvax.Berkeley.EDU (Steve Cousins) Subject: Re: Xerox Dandelion vs. Symbolics? In article <322@mit-amt.MIT.EDU> mob@mit-amt.UUCP writes: >... It is useful and >manipulable but didn't look intended to be programmed by anyone just >off the street. If you want to learn the internals of the user >interface, _then_ i'll grant you four days (and more). > >--Mario O. Bourgoin I think you could argue that *no* machine (AI or otherwise) can be programmed by anyone just off the street :-). I haven't used the Symbolics, but my view of the Dandelion has changed drastically since taking a course on it by Xerox. The interface is very powerful and well-integrated, but the "infant mortality curve" (the time to get good enough not to crash the machines) is somewhat high. [Disclaimer: These machines are supposed to be much better when networked than stand-alone. My change in attitude occurred just as we got ours on the network, and I'm not sure how much to attribute to the class, and how much to attribute to the network]. I like the Dandelion now, but the first 4 days did not give me a good impression of the machine. There is a lot to say about learning a new machine from a guru... Steve Cousins ...ihnp4!wucs!sbc or sbc@wucs.UUCP Washington University ------------------------------ Date: 15 Sep 86 12:58:18 GMT From: clyde!watmath!watnot!watmum!rgatkinson@caip.rutgers.edu (Robert Atkinson) Subject: Re: Xerox Dandelion vs. Symbolics? [vs. Tek 4400 series] In article <580001@hpcnoe.UUCP> jad@hpcnoe.UUCP (John Dilley) writes: >> Why do people choose Symbolics/ZetaLisp/CommonLisp over >> Xerox Dandelion/Interlisp? >> ... >> 3) People are ignorant of what the Dandelion has to offer. > > I have a file of quotes, one of which has to do with this > problem Xerox seems to have. I've heard great things about > Dandelion/Interlisp, and their Smalltalk environments, but have > never seen one of these machines in "real life" (whatever that ... Smalltalk is now (finally!) available from Xerox. An organization known as Parc Place Systems is now licensing both the virtual image and virtual machine implementations for Suns and other workstations. For further info contact: Duane Bay Parc Place Systems 3333 Coyote Hill Rd Palo Alto, CA 94304 -bob ------------------------------ Date: 17 Sep 86 16:49:00 GMT From: princeton!siemens!steve@CAIP.RUTGERS.EDU Subject: Dandelion vs Symbolics I have received enough misinformation about Dandelions and Symbolics machines (by net and mail) that I feel forced to reply. This is not, however, the last word I have to say. I like to keep the net in suspense, acting like I'm saving the BIG REVELATION for later. Key: S= Symbolics, X = Xerox Dandelions - point against X, for S = point for X, against S * misinformation against X, fact in favor (my opinion of course) ? point not classifiable in previous categories A writer who prefers to remain anonymous pointed out: - If your system is bigger than 32 Mb, it can't be done on a Xerox machine. - It takes a great deal of effort to get good numerical performance on X. - X. editor is slow on big functions & lists. My opinion is that it is bad programming style to have such large functions. However, sometimes the application dictates and so this is a point. * "Garbage collection is much more sophisticated on Symbolics" To my knowledge this is absolutely false. S. talks about their garbage collection more, but X's is better. Discuss this more later if you want. * Preference for Zmacs with magic key sequences to load or compile portions of a file, over Dandelion. People who never learn how to use the X system right have this opinion. more later. * "Symbolics system extra tools better integrated" Again, to my knowledge this is false. I know people who say no two tools of S. work together without modification. I have had virtually no trouble with diverse X. tools working together. ? "S. has more tools and functions available e.g. matrix pkg." On the other hand I have heard S. described as a "kitchen sink" system full of many slightly different versions of the same thing. There is a general belief that the reason the X system is around 5 - 6 Mb vs. S. around 24 is that S. includes more tools & packages. + When you load in most of the biggest of the tools & packages to the X system you still are down around 6 - 7 Mb! + If your network is set up reasonably, then it is trivial to load whatever packages you want. It is very nice NOT to have junk cluttering up your system that you don't want. ? "The difference in size reflects how much space you have for CONSes, etc." Huh? I have 20Mb available, yet I find myself actually using less than 7Mb. My world is 7Mb. If I CONS a list 3 Mb long, my world will be 10Mb. Royt@gatech had some "interesting" observations: + Performance per dollar: you can get at least 5 X machines for the cost of a single S machine. AT LEAST. Both types prefer to be on networks with fileservers etc., which adds overhead to both. ? X abysmally slow for baby GPS etc. My guess is that whoever ported/wrote the software didn't know how to get performance out of the X machines. It's not too hard, but it's not always obvious either. = Xerox is getting on the Commonlisp bandwagon only a little late. But how "common" is Commonlisp when window packages are machine dependent? = For every quirk you find in Interlisp (".. Lord, that lisp seems weird to me! I mean, comments that return values??"), I can find one in Commonlisp. (Atoms whose print names are 1.2 for example.) + X has nice windows, less complicated than S. No one i know has ever crashed a X machine by messing with the windows. Opposite for S. machine. + structure editor on X machine, none on S. * "Dandelions *lack* decent file manipulation..." Wrong, comment later. ? he has bad experience with the old IP/TCP package. Me too, but the new one works great. (The X NS protocols actually are quite good but the rest of the world doesn't speak them :-(). ? "..Typically, what I do and what other people do .. is enter a function in the lisp window, which makes it very difficult ..." Didn't you realise you must be doing something wrong? That's not how you enter functions! You give other examples of how you and your cohorts don't know how to use the Xerox system right. You're too stuck on the old C & Fortran kinds of editing and saving stuff. * He goes on about reliability of X being the pits. Every person I have known who learned to use the X machine caused it to crash in various ways, but by the time (s)he had enough experience to be able to explain what he did to someone else, the machine no longer crashed. I guess the X machines have a "novice detector". My understanding is that S has its problems too. One guy had bad experience with KEE, which was developed on X. I do not think his experience is representative. What he did say was that it kept popping up windows he didn't want; X systems make much more use of sophisticated window and graphic oriented tools and interfaces than S, but it doesn't often pop up useless windows in general. Dave@milano thinks S offers reliable hardware, reliable software, and good service that X doesn't. WRONG! At his site, they were obviously doing something sytematically wrong with their machines, and they didn't get a good repairman. I can give you horror stories about Symbolics, too, but I have some pretty reliable points: + At a site I know they have around 20 S. They have sophisticated users and they do their own board swapping. Still they have 10% downtime. + At my site we have very roughly 20 machine-years with X. Total downtime less than 2 machine weeks. + S. has such hardware problems that a) they have a "lemon" program where you can return your machine for a new one, b) their service contracts are OUTRAGEOUSLY EXPENSIVE! These lisp machines are very complex systems. If you don't have someone teach you, who already knows the right ways to use the machine, then it will take you more than 4 months to learn how to use it to the best advantage. Hell, I've been using a Dandelion almost constantly for close to three years and there are still subsystems that I only know superficially, and which I know I could make better use of! If the same isn't true of Symbolics it can only be because the environment is far less rich. It is not difficult to learn these subsystems; the problem is there's just SO MUCH to learn. Interlisp documentation was just re-done and it's 4.5 inches thick! (Used to be only 2.25) Finally, I will expound a little on why Xerox is better than Symbolics. The Xerox file system and edit/debug cycle is far superior to an old- fashioned standard system like Symbolics which has a character-oriented editor like Zmacs. The hard part for many people to learn the Xerox file system, is that first they have to forget what they know about editors and files. A lot of people are religious about their editors, so this step can be nearly impossible. Secondly, the documentation until the new version was suitable primarily for people who already knew what was going on. That hurt a lot. (It took me maybe 1.5 years before I really got control of the file package, but I was trying to learn Lisp in the first place, and everything else at the same time.) Now it's much much faster to learn. The old notion of files and editors is like assembly language. Zmacs with magic key sequences to compile regions etc. is like a modern, good assembler with powerful macros and data structures and so forth. Xerox's file system is like Fortran/Pascal/C. Ask the modern assembly programmer what he sees in Fortranetc. and he'll say "nothing". It'll be hard for him to learn. He's used to the finer grain of control over the machine that assembly gives him and he doesn't understand how to take advantage of the higher level features of the Fortranetc. language. Before you flame at me too much, remember I am analogizing to a modern, powerful assembler not the trash you used 5 years ago on your TRS-80. The xerox file package treats a file as a database of function definitions, variable values, etc. and gives you plenty of power to deal with them as databases. This note is long enough and I don't know what else to say so I'll drop this topic somewhat unfinished (but I will NOT give lessons on how to use the Xerox file package). A final final note: the guy down the hall from me has used S. for some years and now has to learn X. He isn't complaining too much. I hope he'll post his own remarks soon, but I've got to relate one story. I wanted to show him something, and of course when I went to run it it didn't work right. As I spent a minute or two eradicating the bug, he was impressed by the use of display-oriented tools on the Dandelion. He said, "Symbolics can't even come close." Steven J. Clark, Siemens RTL {ihnp4!princeton or topaz}!siemens!steve ------------------------------ End of AIList Digest ******************** From csnet_gateway Sun Sep 21 18:42:51 1986 Date: Sun, 21 Sep 86 18:42:43 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #192 Status: RO AIList Digest Sunday, 21 Sep 1986 Volume 4 : Issue 192 Today's Topics: Conferences - AI and Law & Logic in Computer Science & SIGIR R&D in Information Retrieval & Logical Solutions to the Frame Problem & CSCW '86 Program ---------------------------------------------------------------------- Date: 13 Aug 86 20:36:33 EDT From: MCCARTY@RED.RUTGERS.EDU Subject: Conference - AI and Law CALL FOR PAPERS: First International Conference on ARTIFICIAL INTELLIGENCE AND LAW May 27-29, 1987 Northeastern University Boston, Massachusetts, USA In recent years there has been an increased interest in the applications of artificial intelligence to law. Some of this interest is due to the potential practical applications: A number of researchers are developing legal expert systems, intended as an aid to lawyers and judges; other researchers are developing conceptual legal retrieval systems, intended as a complement to the existing full-text legal retrieval systems. But the problems in this field are very difficult. The natural language of the law is exceedingly complex, and it is grounded in the fundamental patterns of human common sense reasoning. Thus, many researchers have also adopted the law as an ideal problem domain in which to tackle some of the basic theoretical issues in AI: the representation of common sense concepts; the process of reasoning with concrete examples; the construction and use of analogies; etc. There is reason to believe that a thorough interdisciplinary approach to these problems will have significance for both fields, with both practical and theoretical benefits. The purpose of this First International Conference on Artificial Intelligence and Law is to stimulate further collaboration between AI researchers and lawyers, and to provide a forum for the latest research results in the field. The conference is sponsored by the Center for Law and Computer Science at Northeastern University. The General Chair is: Carole D. Hafner, College of Computer Science, Northeastern University, 360 Huntington Avenue, Boston MA 02115, USA; (617) 437-5116 or (617) 437-2462; hafner.northeastern@csnet-relay. Authors are invited to contribute papers on the following topics: - Legal Expert Systems - Conceptual Legal Retrieval Systems - Automatic Processing of Natural Legal Texts - Computational Models of Legal Reasoning In addition, papers on the relevant theoretical issues in AI are also invited, if the relationship to the law can be clearly demonstrated. It is important that authors identify the original contributions presented in their papers, and that they include a comparison with previous work. Each submission will be reviewed by at least three members of the Program Committee (listed below), and judged as to its originality, quality and significance. Authors should submit six (6) copies of an Extended Abstract (6 to 8 pages) by January 15, 1987, to the Program Chair: L. Thorne McCarty, Department of Computer Science, Rutgers University, New Brunswick NJ 08903, USA; (201) 932-2657; mccarty@rutgers.arpa. Notification of acceptance or rejection will be sent out by March 1, 1987. Final camera-ready copy of the complete paper (up to 15 pages) will be due by April 15, 1987. Conference Chair: Carole D. Hafner Northeastern University Program Chair: L. Thorne McCarty Rutgers University Program Committee: Donald H. Berman Northeastern University Michael G. Dyer UCLA Edwina L. Rissland University of Massachusetts Marek J. Sergot Imperial College, London Donald A. Waterman The RAND Corporation ------------------------------ Date: Tue, 9 Sep 86 09:26:57 PDT From: Moshe Vardi Subject: Conference - Logic in Computer Science CALL FOR PAPERS SECOND ANNUAL SYMPOSIUM ON LOGIC IN COMPUTER SCIENCE 22 - 25 June 1987 Cornell University, Ithaca, New York, USA THE SYMPOSIUM will cover a wide range of theoretical and practical issues in Computer Science that relate to logic in a broad sense, including algebraic and topological approaches. Suggested (but not exclusive) topics of interest include: abstract data types, computer theorem proving, verification, concurrency, type theory and constructive mathematics, data base theory, foundations of logic programming, program logics and semantics, knowledge and belief, software specifications, logic-based programming languages, logic in complexity theory. Organizing Committee K. Barwise E. Engeler A. Meyer W. Bledsoe J. Goguen R. Parikh A. Chandra (chair) D. Kozen G. Plotkin E. Dijkstra Z. Manna D. Scott Program Committee S. Brookes D. Gries (chair) J.-P. Jouannaud A. Nerode L. Cardelli J. Goguen R. Ladner G. Plotkin R. Constable Y. Gurevich V. Lifschitz A. Pnueli M. Fitting D. Harel G. Longo P. Scott PAPER SUBMISSION. Authors should send 16 copies of a detailed abstract (not a full paper) by 9 DECEMBER 1986 to the program chairman: David Gries -- LICS (607) 255-9207 Department of Computer Science gries@gvax.cs.cornell.edu Cornell University Ithaca, New York 14853 Abstracts must be clearly written and provide sufficient detail to allow the program committee to assess the merits of the paper. References and comparisons with related work should be included where appropriate. Abstracts must be no more than 2500 words. Late abstracts or abstracts departing significantly from these guidelines run a high risk of not being considered. If a copier is not available to the author, a single copy of the abstract will be accepted. Authors will be notified of acceptance or rejection by 30 JANUARY 1987. Accepted papers, typed on special forms for inclusion in the symposium proceedings, will be due 30 MARCH 1987. The symposium is sponsored by the IEEE Computer Society, Technical Committee on Mathematical Foundations of Computing and Cornell University, in cooperation with ACM SIGACT, ASL, and EATCS. GENERAL CHAIRMAN LOCAL ARRANGEMENTS Ashok K. Chandra Dexter C. Kozen IBM Thomas J. Watson Research Center Department of Computer Science P.O. Box 218 Cornell University Yorktown Heights, New York 10598 Ithaca, New York 14853 (914) 945-1752 (607) 255-9209 ashok@ibm.com kozen@gvax.cs.cornell.edu ------------------------------ Date: Tue, 12 Aug 86 16:16:01 cdt From: Don Subject: Conference - SIGIR Conf. on R&D in Information Retrieval Association for Computing Machinery (ACM) Special Interest Group on Information Retrieval (SIGIR) 1987 International Conference on Research and Development in Information Retrieval June 3-5, 1987 Monteleone Hotel (in the French Quarter) New Orleans, Louisiana USA CALL FOR PAPERS Papers are invited on theory, methodology, and applications of information retrieval. Emerging areas related to infor- mation retrieval, such as office automation, computer hardware technology, and artificial intelligence and natural language processing are welcome. Topics include, but are not limited to: retrieval system modeling user interfaces retrieval in office environments mathematical models system development and evaluation natural language processing knowledge representation linguistic models hardware development complexity problems multimedia retrieval storage and search techniques cognitive and semantic models retrieval system performance information retrieval and database management Submitted papers can be either full length papers of approx- imately twenty to twenty-five pages or extended abstracts of no more than ten pages. All papers should contain the authors' contributions in comparison to existing solutions to the same or to similar problems. Important Dates Submission Deadline December 15, 1986 Acceptance Notification February 15, 1987 Final Copy Due March 20, 1987 Conference June 3-5, 1987 Four copies of each paper should be submitted. Papers sub- mitted from North America can be sent to Clement T. Yu; sub- missions from outside North America should be sent to C. J. "Keith" van Rijsbergen. Conference Chairman Program Co-Chairmen Donald H. Kraft Clement T. Yu C. J. "Keith" van Rijsbergen Department of Department of Department of Computer Science Electrical Engineering Computer Science Louisiana State University and Computer Science University of Glascow Baton Rouge, LA 70803 University of Illinois, Lilybank Gardens Chicago Glascow G12 8QQ Chicago, IL 60680 SCOTLAND (504) 388-1495 (312) 996-2318 (041) 339-8855 For details, contact the Conference Chairman at kraft%lsu@csnet-relay or Michael Stinson, the Arrangements Chairman at stinson%lsu@csnet-relay. Don Kraft kraft%lsu@csnet-relay ------------------------------ Date: Fri, 19 Sep 86 16:04:04 CDT From: Glenn Veach Subject: Conference - Logical Solutions to the Frame Problem CALL FOR PAPERS WORKSHOP ON LOGICAL SOLUTIONS TO THE FRAME PROBLEM The American Association for Artificial Intelligence (AAAI) is sponsoring this workshop in Lawrence, Kansas from March 23 to March 25,1987. The frame problem is one of the most fundamental problems in Artificial Intelligence and essentially is the problem of describing in a computationally reasonable manner what properties persist and what properties change as action are performed. The intrinsic problem lies in the fact that we cannot expect to be able to exhaustively list for every possible action (or combination of concurrent actions) and for every possible state of the world how that action (or concurrent actions) change the truth or falsity of each individual fact. We can only list the obvious results of the action and hope that our basic inferential system will be able to deduce the truth or falsity of the other less obvious facts. In recent years there have been a number of approaches to constructing new kinds of logical systems such as non-monotonic logics, default logics, circumscription logics, modal reflexive logics, and persistence logics which hopefully can be applied to solving the frame problem by allowing the missing facts to be deduced. This workshop will attempt to bring together the proponents of these various approaches. Papers on logics applicable to the problem of reasoning about such unintended consequences of actions are invited for consideration. Two copies of either an extended abstract or a full length paper should be sent to the workshop chairman before Nov 20,1986. Acceptance notices will be mailed by December 1,1986 along with instructions for preparing the final versions of accepted papers. The final versions are due January 12,1987. In order to encourage vigorous interaction and exchange of ideas the workshop will be kept small -- about 25 participants. There will be individual presentations and ample time for technical discussions. An attempt will be made to define the current state of the art and future research needs. Partial travel support (from AAAI) for participants is available. Workshop Chairman: Dr. Frank M. Brown Dept Computer Science 110 strong Hall The University of Kansas Lawrence, Kansas (913) 864-4482 Please send any net inquiries to: veach@ukans.csnet ------------------------------ Date: Tue 2 Sep 86 15:20:55-EDT From: Irene Greif Subject: Conference - CSCW '86 Program Following is the program for CSCW '86: the Conference on Computer-Supported Cooperative Work . Registration material can be obtained from Barbara Smith at MCC (basmith@mcc). [Contact the author for the full program. -- KIL] ------------------------------ End of AIList Digest ******************** From csnet_gateway Sun Sep 21 18:43:06 1986 Date: Sun, 21 Sep 86 18:42:56 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #193 Status: RO AIList Digest Sunday, 21 Sep 1986 Volume 4 : Issue 193 Today's Topics: Seminars - Backtrach Search for Constraint Satisfaction (SRI) & Minisupercomputers and AI Machines (CMU) & Equal Opportunity Interactive Systems (SU), Seminars (Past) - AI in Communication Networks (Rutgers) & Goal Integration in Heuristic Algorithm Design (Rutgers) & Long-Term Planning Systems (TI) & Learning by Understanding Analogies (SRI) & Belief Revision (SRI) & Factorization in Experiment Generation (SRI) ---------------------------------------------------------------------- Date: Wed 17 Sep 86 13:39:03-PDT From: Amy Lansky Subject: Seminar - Backtrach Search for Constraint Satisfaction (SRI) IMPROVING BACKTRACK SEARCH ALGORITHMS FOR CONSTRAINT-SATISFACTION PROBLEMS Rina Dechter (DECHTER@CS.UCLA.EDU) Cognitive System Laboratory, Computer Science Department, U.C.L.A. and Artificial Intelligence Center, Hughes Aircraft Company 11:00 AM, TUESDAY, September 23 SRI International, Building E, Room EJ228 The subject of improving search efficiency has been on the agenda of researchers in the area of Constraint-Satisfaction- Problems (CSPs) for quite some time. A recent increase of interest in this subject, concentrating on backtrack search, can be attributed to its use as the control strategy in PROLOG, and in Truth-Maintenance-Systems (TMS). The terms ``intelligent backtracking'', ``selective backtracking'', and ``dependency- directed backtracking'' describe various efforts for producing improved dialects of backtrack search in these systems. In this talk I will review the common features of these attempts and will present two schemes for enhancing backtrack search in solving CSPs. The first scheme, a version of "look-back", guides the decision of what to do in dead-end situations. Specifically, we concentrate on the idea of constraint recording, namely, analyzing and storing the reasons for the dead-ends, and using them to guide future decisions, so that the same conflicts will not arise again. We view constraint recording as a process of learning, and examine several possible learning schemes studying the tradeoffs between the amount of learning and the improvement in search efficiency. The second improvement scheme exploits the fact that CSPs whose constraint graph is a tree can be solved easily, i.e., in linear time. This leads to the following observation: If, in the course of a backtrack search, the subgraph resulting from removing all nodes corresponding to the instantiated variables is a tree, then the rest of the search can be completed in linear time. Consequently, the aim of ordering the variables should be to instantiate as quickly as possible a set of variables that cut all cycles in the constraint-graph (cycle-cutset). This use of cycle-cutsets can be incorporated in any given "intelligent" backtrack and is guaranteed to improve it (subject to minor qualifications). The performance of these two schemes is evaluated both theoretically and experimentally using randomly generated problems as well as several "classical" problems described in the literature. VISITORS: Please arrive 5 minutes early so that you can be escorted up from the E-building receptionist's desk. Thanks! ALSO: NOTE DAY CHANGE!!! (Tuesday -- this week only) ------------------------------ Date: 17 Sep 86 14:53:24 EDT From: Barbara.Grandillo@n.sp.cs.cmu.edu Subject: Seminar - Minisupercomputers and AI Machines (CMU) Special Computer Science Seminar Speaker: Professor Kai Hwang University of Southern California Title: Design Issues of Minisupercomputers and AI Machines Date: Monday, September 22, 1986 Time: 12:00 noon Place: Wean Hall 4605 In this seminar, Dr. Hwang will address the fundamental issues in designing efficient multiprocessor/multicomputer minisupercomputers or AI machines. The talk covers the systems architectural choices, interprocessor communication mechanisms, resource allocation methods, I/O and OS functions, mapping of parallel algorithms, and the creation of parallel programming environment for these machines. These design issues and their possible solutions are drawn from the following commercial or exploratory systems: Alliant FX/8, FPS T-Series and M64 Series,Flex/32, Encore Multimax, Flex/32, Elxsi 6400, Sequent 8000, Connection Machine, BBN Butterfly, FAIM-1, Alice, Dado, Soar, and Redfiflow, etc. Dr. Hwang will also assess the technological basis and future trends in low-cost supercomputing and AI processing. ------------------------------ Date: 19 Sep 86 0845 PDT From: Rosemary Napier Subject: Seminar - Equal Opportunity Interactive Systems (SU) Computer Science Colloquium Tuesday, October 7, 1986, 4:15PM, Terman Auditorium "Equal Opportunity Interactive Systems and Innovative Design" Harold Thimbleby Dept. of Computer Science University of York Heslington, York United Kingdom YO1 5DD Most interactive systems distinguish between the input and output of information. Equal opportunity is a design heuristic that discards these distinctions; it was inspired by polymodality in logic programming and a well-known problem solving heuristic. The seminar makes the case for equal opportunity, and shows how several user engineering principles, techniques and systems can be reappraised under equal opportunity. By way of illustration, equal opportunity is used to guide the design of a calculator and spreadsheet. The resulting systems have declarative user interfaces and are arguably easier to use despite complex operational models. About the speaker: Harold Thimbleby did his doctoral research in user interface design. He joined the Computer Science department at York in 1982 and is currently on sabbatical at the Knowledge Sciences Institute, Calgary. He is currently writing a book on the application of formal methods as heuristics for user interface design. ------------------------------ Date: 8 Sep 86 23:50:47 EDT From: Tom Fawcett Subject: Seminar - AI in Communication Networks (Rutgers) The first speaker of this year's Machine Learning Seminar series at Rutgers will be Andrew Jennings of Telecom Australia, speaking on "AI in Communication Networks". Dr. Andrews will speak in Hill-423 at 11 AM on THURSDAY, September 18th (NB: this is NOT the standard day for the ML series). The abstract follows: Andrew Jennings (Arpanet address: munnari!trlamct.oz!andrew@seismo.CSS.GOV) Telecom Australia AI in Communication Networks Expert systems are gaining wide application in communication systems, especially in the areas of maintenance, design and planning. Where there are large bodies of existing expertise, expert systems are a useful programming technology for capturing and making use of that expertise. However will AI techniques be limited to retrospective capturing of expertise or can they be of use for next generation communication systems? This talk will present several projects that aim to make use of AI techniques in next-generation communication networks. An important aspect of these systems is their ability to learn from experience. This talk will discuss some of the difficulties in developing learning in practical problem domains, and the value of addressing these difficulties now. In particular the problems of learning in intractable problem domains is of great importance for these problems and some ongoing work on this problem will be presented. The projects discussed include a system for capacity assignment in networks, a project to develop AI systems for routing in fast packet networks and a system for VLSI design from a high level specification. ------------------------------ Date: 9 Sep 86 12:43:20 EDT From: Tom Fawcett Subject: Seminar - Goal Integration in Heuristic Algorithm Design (Rutgers) Next week, on Tuesday, September 16th in Hill 423 at 11 AM, Jack Mostow will give a talk based on his work with Kerstin Voigt, entitled "A Case Study of Goal Integration in Heuristic Algorithm Design". This a joint ML/III seminar, and is a dry run for a talk being given at the Knowledge Compilation Workshop. There's no paper for the talk, but Jack recommends his AAAI86 article with Bill Swartout as good background reading. The abstract follows: Jack Mostow Rutgers University (Arpanet address: MOSTOW@RED.RUTGERS.EDU) A Case Study of Goal Integration in Heuristic Algorithm Design: A Transformational Rederivation of MYCIN's Therapy Selection Algorithm An important but little-studied aspect of compiling knowledge into efficient procedures has to do with integrating multiple, sometimes conflicting goals expressed as part of that knowledge. We are developing an artificial intelligence model of heuristic algorithm design that makes explicit the interactions among multiple goals. The model will represent intermediate states and goals in the design process, transformations that get from one state to the next, and control mechanisms that govern the selection of which transformation to apply next. It will explicitly model the multiple goals that motivate and are affected by each design choice. We are currently testing and refining the model by using it to explain the design of the algorithm used for therapy selection in the medical expert system MYCIN. Previously we analyzed how this algorithm derives from the informal specification "Find the set of drugs that best satisfies the medical goals of maximizing effectiveness, minimizing number of drugs, giving priority to treating likelier organisms, [etcetera]." The reformulation and integration of these goals is discussed in Mostow & Swartout's AAAI86 paper. Doctoral student Kerstin Voigt is implementing a complete derivation that will address additional goals important in the design of the algorithm, such as efficient use of time, space, and experts. ------------------------------ Date: Mon 18 Aug 86 16:28:02-CDT From: Rajini Subject: Seminar - Long-Term Planning Systems (TI) Dr. Jim Hendler, Assistant Professor at Univ of Maryland, is a giving a special seminar at 10:00 am on August 28th. Abstract of his talk follows. It will be held in Conference room #2, Computer Science Center, Texas Instruments, Dallas. --Rajini rajini@ti-csl (214) 995-0779 Long-term planning systems James Hendler Computer Science Dept. University of Maryland College Park, Md. 20903 Most present day planning systems work in domains where a single goal is planned for a single user. Further, the only object changing the world is the planner itself. The few systems that go beyond this, for example Vere's DEVISER system, tend to work in domains where the world, although changing, behaves according to a set of well-defined rules. In this talk we describe on-going research directed at extending planning systems to function in the dynamic environments necessary for such tasks as job-shop scheduling, process control, and autonomous vehicle missions. The talk starts by describing the inadequacies of present-day systems for working in such tasks. We focus on two, necessity of a static domain and inability to handle large numbers of interacting goals, and show some of the extensions needed to handle these systems. We describe an extension to marker-passing, a parallel, spreading activation system, which can be used for handling the goal interaction problems, and we discuss representational issues necessary to handling dynamic worlds. We end by describing work on a system which is being implemented to deal with these problems. ------------------------------ Date: Tue 19 Aug 86 19:55:33-PDT From: Margaret Olender Subject: Seminar - Learning by Understanding Analogies (SRI) Russell Greiner, Toronto, will be guest speaker at the RR Group's PlanLunch (August 20, EJ228, 11:00am). LEARNING BY UNDERSTANDING ANALOGIES This research describes a method for learning by analogy---i.e., for proposing new conjectures about a target analogue based on facts known about a source analogue. After formally defining this process, we present heuristics which efficiently guide it to the conjectures which can help solve a given problem. These rules are based on the view that a useful analogy is one which provides the information needed to solve the problem, and no more. Experimental data confirms the effectiveness of this approach. ------------------------------ Date: Wed 20 Aug 86 16:02:46-PDT From: Amy Lansky Subject: Seminar - Belief Revision (SRI) IS BELIEF REVISION HARDER THAN YOU THOUGHT? Marianne Winslett (WINSLETT@SCORE) Stanford University, Computer Science Department 11:00 AM, MONDAY, Aug. 25 SRI International, Building E, Room EJ228 Suppose one wishes to construct, use, and maintain a database of knowledge about the real world, even though the facts about that world are only partially known. In the AI domain, this problem arises when an agent has a base set of extensional beliefs that reflect partial knowledge about the world, and then tries to incorporate new, possibly contradictory extensional knowledge into the old set of beliefs. We choose to represent such an extensional knowledge base as a logical theory, and view the models of the theory as possible states of the world that are consistent with the agent's extensional beliefs. How can new information be incorporated into the extensional knowledge base? For example, given the new information that "b or c is true," how can we get rid of all outdated information about b and c, add the new information, and yet in the process not disturb any other extensional information in the extensional knowledge base? The burden may be placed on the user or other omniscient authority to determine exactly which changes in the theory will bring about the desired set of models. But what's really needed is a way to specify the update intensionally, by stating some well-formed formula that the state of the world is now known to satisfy and letting internal knowledge base mechanisms automatically figure out how to accomplish that update. In this talk we present semantics and algorithms for an operation to add new information to extensional knowledge bases, and demonstrate that this action of extensional belief revision is separate from, and in practice must occur prior to, the traditional belief revision processes associated with truth maintenance systems. ------------------------------ Date: Wed 3 Sep 86 14:51:36-PDT From: Amy Lansky Subject: Seminar - Factorization in Experiment Generation (SRI) FACTORIZATION IN EXPERIMENT GENERATION Devika Subramanian Stanford University, Computer Science Department 11:00 AM, MONDAY, September 8 SRI International, Building E, Room EJ228 Experiment Generation is an important part of incremental concept learning. One basic function of experimentation is to gather data to refine an existing space of hypotheses. In this talk, we examine the class of experiments that accomplish this, called discrimination experiments, and propose factoring as a technique for generating them efficiently. ------------------------------ End of AIList Digest ******************** From csnet_gateway Sun Sep 21 06:59:00 1986 Date: Sun, 21 Sep 86 06:58:55 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #194 Status: RO AIList Digest Sunday, 21 Sep 1986 Volume 4 : Issue 194 Today's Topics: Seminars (Past) - Rule Induction in Computer Chess (ACM LA Chapter) & Mechanization of Geometry (SU) & Automatic Algorithm Designer (CMU) & Representations and Checkerboards (CMU) & Deriving Problem Reduction Operators (Rutgers) & Evolution of Automata (SRI) & Active Reduction of Uncertainty (UPenn) & Rational Conservatism and the Will to Believe (CMU) & BiggerTalk: An Object-Oriented Extension to Prolog (UTexas) ---------------------------------------------------------------------- Date: 21 Aug 86 12:01:50 PDT (Thu) From: ledoux@aerospace.ARPA Subject: Seminar - Rule Induction in Computer Chess (ACM LA Chapter) ACM LOS ANGELES CHAPTER DINNER MEETING WEDNESDAY, 3 SEPTEMBER 1986 STRUCTURED EXPERT RULE INDUCTION Expert Systems and Computer Chess Speaker: Dr. Alen Shapiro One of the major problems with expert systems is "the knowledge engineering bottleneck." This occurs when development is delayed because specifications are unavailable and either the expert system developers need time to learn the problem, or else the domain experts who already know the problem need time to learn how to use the often opaque expert system development languages. A promising approach to overcoming the bottleneck is to build tools that automatically extract knowledge from the domain experts. This talk presents an overview of inductive knowledge acquisition and the results of experiments in inductive rule generation in the domain of chess endgames. The system that will be described was able to generate humanly-understandable rules and to play correct chess endgames. This research has significant implications for the design of expert system languages and rule induction programs. The talk is also an interesting look into the world of computer chess. Dr. Shapiro, a Fellow of the Turing Institute since its inception in 1983, received his Ph.D. in Machine Intelligence from the University of Edinburgh in 1983. From 1979 to 1986 he was associated with Intelligent Terminals, Ltd., and a member of the Rulemaster and Expert-Ease design teams. He has served as Visiting Professor at the University of Illinois on two occasions. His publications include articles on pattern recognition, automatic induction of chess classification rules, and (with David Michie), "A Self-Commenting Facility for Inductively Synthesized Endgame Expertise." In 1986 Dr. Shapiro joined the New Technology Department at Citicorp-TTI in Santa Monica as a Computer Scientist concerned with the development of inductive knowledge engineering tools for the banking industry. PLACE Amfac Hotel 8601 Lincoln Blvd. corner of Lincoln & Manchester Westchester, California 8:00 p.m. ------------------------------ Date: Mon, 18 Aug 86 11:25:38 PDT From: coraki!pratt@Sun.COM (Vaughan Pratt) Subject: Seminar - Mechanization of Geometry (SU) SPEAKER Professor Wu Wen-tsun TITLE Mechanization of Geometry DATE Thursday, August 21 TIME 2:00 pm PLACE Margaret Jacks Hall, room 352 A mechanical method of geometry based on Ritt's characteristic set theory will be described which has a variety of applications including mechanical geometry theorem proving in particular. The method has been implemented on computers by several researchers and turns out to be efficient for many applications. BACKGROUND Professor Wu received his doctorate in France in the 1950's, and was a member of the Bourbaki group. In the first National Science and Technology Awards in China in 1956, Professor Wu was one of three people awarded a first prize for their contributions to science and technology. He is currently the president of the Chinese Mathematical Society. In 1977, Wu extended classical algebraic geometry work of Ritt to an algorithm for proving theorems of elementary geometry. The method has recently become well-known in the Automated Theorem Proving community; at the University of Texas it has been applied it to the machine proof of more than 300 theorems of Euclidean and non-Euclidean geometry. ------------------------------ Date: 5 September 1986 1527-EDT From: Betsy Herk@A.CS.CMU.EDU Subject: Seminar - Automatic Algorithm Designer (CMU) Speaker: David Steier Date: Friday, Sept. 12 Place: 5409 Wean Hall Time: 3:30 p.m. Title: Integrating multiple sources of knowledge in an automatic algorithm designer One of the reasons that designing algorithms is so difficult is the large amount of knowledge needed to guide the design process. In this proposal, I identify nine sources of such knowledge within four general areas: general problem-solving, algorithm design and implementation techniques, knowledge of the application domain, and methods for learning from experience. To understand how knowledge from these sources can be represented and integrated, I propose to build a system that automatically designs algorithms. An implementation of the system, Designer-Soar, uses several of the knowledge sources described in the proposal to design several very simple algorithms. The goal of the thesis is to extend Designer-Soar to design moderately complex algorithms in a domain such as graph theory or computational geometry. ------------------------------ Date: 10 September 1986 1019-EDT From: Elaine Atkinson@A.CS.CMU.EDU Subject: Seminar - Representations and Checkerboards (CMU) SPEAKER: Craig Kaplan, CMU, Psychology Department TITLE: "Representations and Checkerboards" DATE: Thursday, September 11 TIME: 4:00 p.m. PLACE: Adamson Wing, BH Given the right representation, tricky "insight" problems often become trivial to solve. How do people arrive at the right representations? What factors affect people's ability to shift representations, and how can understanding these factors help us understand why insight problems are so difficult? Evidence from studies using the Mutilated Checkerboard Problem points to Heuristic Search as a powerful way of addressing these questions. Specifically, it suggest that the quality of the match between people's readily available search heuristics and problem characteristics is a major determinant of problem difficulty for some problems. ------------------------------ Date: 11 Sep 86 20:01:20 EDT From: RIDDLE@RED.RUTGERS.EDU Subject: Seminar - Deriving Problem Reduction Operators (Rutgers) I am giving a practice talk of a talk I will be giving in a few weeks. It is at 1 pm in 423 on Monday the 15th. Everyone is invited and all comments are welcome. The abstract follows. This research deals with automatically shifting from one problem representation to another representation which is more efficient, with respect to a given problem solving method, for this problem class. I attempt to discover general purpose primitive representation shifts and techniques for automating them. To achieve this goal, I am defining and automating the primitive representation shifts explored by Amarel in the Missionaries & Cannibals problem @cite(amarel1). The techniques for shifting representations which I have already defined are: compiling constraints, removing irrelevant information, removing redundant information, deriving macro-operators, deriving problem reduction operators, and deriving macro-objects. In this paper, I will concentrate on the technique for deriving problem reduction operators (i.e., critical reduction) and a method for automating this technique (i.e., invariant reduction). A set of sufficient conditions for the applicability of this technique over a problem class is discussed; the proofs appear in a forthcoming Rutgers technical report. ------------------------------ Date: Wed 10 Sep 86 15:00:22-PDT From: Amy Lansky Subject: Seminar - Evolution of Automata (SRI) THE EVOLUTION OF COMPUTATIONAL CAPABILITIES IN POPULATIONS OF COMPETING AUTOMATA Aviv Bergman (BERGMAN@SRI-AI) SRI International and Michel Kerszberg IFF der KFA Julich, W.-Germany 10:30 AM, MONDAY, September 15 SRI International, Building E, Room EJ228 The diversity of the living world has been shaped, it is believed, by Darwinian selection acting on random mutations. In the present work, we study the emergence of nontrivial computational capabilities in automata competing against each other in an environment where possession of such capabilities is an advantage. The automata are simple cellular computers with a certain number of parameters - characterizing the "Statistical Distribution" of the connections - initially set at random. Each generation of machines is subjected to a test necessitating some computational task to be performed, e.g recognize whether two patterns presented are or are not translated versions of each other. "Adaptive Selection" is used during the task in order to "Eliminate" redundant connections. According to its grade, each machine either dies or "reproduces", i.e. it creates an additional machine with parameters almost similar to its own. The population, it turns out, quickly learns to perform certain tests. When the successful automata are "autopsied", it appears that they do not all complete the task in the same way; certain groups of cells are more active then others, and certain connections have grown or decayed preferentially, but these features may vary from individual to individual. We try to draw some general conclusions regarding the design of artificial intelligence systems, and the understanding of biological computation. We also contrast this approach with the usual Monte-Carlo procedure. ------------------------------ Date: Wed, 13 Aug 86 08:51 EDT From: Tim Finin Subject: Seminar - Active Reduction of Uncertainty (UPenn) Active Reduction of Uncertainty in Multi-sensor Systems Ph.D. Thesis Proposal Greg Hager (greg@upenn-grasp) General Robots and Active Sensory Perception Laboratory University of Pennsylvania Department of Computer and Information Sciences Philadelphia, PA 19104 10:00 AM, August 15, 1986 Room 554 Moore If robots are to perform tasks in unconstrained environments, they will have to rely on sensor information to make decisions. In general, sensor information has some uncertainty associated with it. The uncertainty can be conceptually divided into two types: statistical uncertainty due to signal noise, and incompleteness of information due to limitations of sensor scope. Inevitably, the information needed for proper action will be uncertain. In these cases, the robot will need to take action explicitly devoted to reducing uncertainty. The problem of reducing uncertainty can be studied within the theoretical framework of team decision theory. Team decision theory considers a number of decision makers observing the world via information structures, and taking action dictated by decision rules. Decision rules are evaluated relative to team and individual utility considerations. In this vocabulary, sensors are considered as controllable information structures whose behavior is determined by individual and group utilities. For the problem of reducing uncertainty, these utilities are based on the information expected as the result of taking action. In general, a robot does not only consider direct sensor observations, but also evaluates and combines that data over time relative to some model of the observed environment. In this proposal, information aggregation is modeled via belief systems as studied in philosophy. Reducing uncertainty corresponds to driving the belief system into one of a set of information states. Within this context, the issues that will be addressed are the specification of utilities in terms of belief states, the organization of a sensor system, and the evaluation of decision rules. These questions will first be studied through theory and simulation, and finally applied to an existing multi-sensor system. Advisor: Dr. Max Mintz Committee: Dr. Ruzena Bajcsy (Chairperson) Dr. Zolton Domotor (Philosophy Dept.) Dr. Richard Paul Dr. Stanley Rosenschein (SRI International and CSLI) ------------------------------ Date: 10 Sep 1986 0848-EDT From: Lydia Defilippo Subject: Seminar - Rational Conservatism and the Will to Believe (CMU) CMU PHILOSOPHY COLLOQUIUM JON DOYLE RATIONAL CONSERVATISM AND THE WILL TO BELIEVE DATE: MONDAY SEPTEMBER 15 TIME: 4:OO P.M. PLACE: PORTER HALL, RM 223d * Much of the reasoning automated in artificial intelligence is either mindless deductive inference or is intentionally non-deductive. The common explanations of these techniques, when given, are not very satisfactory, for the real explanations involve the notion of bounded rationality, while over time the notion of rationality has been largely dropped from the vocabulary of artificial intelligence. We present the notion of rational self-government, in which the agent rationally guides its own limited reasoning to whatever degree is possible, via the examples of rational conservatism and rationally adopted assumptions. These ideas offer improvements on the practice of mindless deductive inference and explantions of some of the usual non-deductive inferences. ------------------------------ Date: Mon 15 Sep 86 10:35:02-CDT From: ICS.BROWNE@R20.UTEXAS.EDU Subject: Seminar - BiggerTalk: An Object-Oriented Extension to Prolog (UTexas) Object-Oriented Programming Meeting Friday, September 19 2:00-3:00 p.m. Taylor 3.128 BiggerTalk: An Object-Oriented Extension to Prolog Speaker: Eric Gullichsen MCC Software Technology Program BiggerTalk is a system of Prolog routines which provide a capability for object-oriented programming in Prolog. When compiled into a standard Prolog environment, the BiggerTalk system permits programming in the object-oriented style of message passing between objects, themselves defined as components of a poset (the 'inheritance structure') created through other BiggerTalk commands. Multiple inheritance of methods and instance variables is provided dynamically. The full functional capability of Prolog is retained, and Prolog predicates can be invoked from within BiggerTalk methods. A provision exists for storage of BiggerTalk objects in the MCC-STP Object Server, a shared permanent object repository. The common external form for objects in the Server permits (restricted) sharing of objects between BiggerTalk and Zetalisp Flavors, the two languages currently supported by the Server. Concurrent access to permanent objects is mediated by the server. This talk will discuss a number of theoretical and pragmatic issues of concern to BiggerTalk and its interface to the Object Server. Some acquaintance with the concepts of logic programming and object-oriented programming will be assumed. ------------------------------ End of AIList Digest ******************** From csnet_gateway Thu Sep 25 06:41:54 1986 Date: Thu, 25 Sep 86 06:41:49 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #195 Status: R AIList Digest Thursday, 25 Sep 1986 Volume 4 : Issue 195 Today's Topics: Queries - Public-Domain Ops5 & XLisp & Lsmalltalk & Kyoto Common Lisp & LISP-to-FORTRAN Conversion & Cognitive Science Schools, AI Tools - OPS5 & OPSx & UNIX Tools, Expert Systems - Literature Resources & Implementation Styles ---------------------------------------------------------------------- Date: 20 Sep 86 15:32:14 GMT From: ritcv!eer@ROCHESTER.ARPA (Ed Reed) Subject: Public domain Ops5 in any language I'm looking for one of the versions of OPS5 in lisp (or ?) that's in the public domain. I've heard that there are pd versions running around, but haven't found any here, yet. If in lisp (as I expect) I can use FranzLisp, DecCommonLisp, and xlisp, and maybe InterLisp on a Xerox Dlion. Thanks for the help.. Ed Reed Rochester Inst. Technology, Rochester, NY ....seismo!rochester!ritcv Delphi: eertest GEnie: SQA.INC ------------------------------ Date: 19 Sep 1986 21:30-EDT From: cross@wpafb-afita Subject: xlisp query Would appreciate a pointer to where I could download the source code for xlisp 1.6 and any demonstratable programs written in xlisp. I'm aware of the stuff published in AI Expert and have downloaded it, but cannot find the source code. Thanks in advance. Steve Cross ------------------------------ Date: 24 Sep 86 03:50:21 GMT From: booter@lll-crg.arpa (Elaine Richards) Subject: Lsmalltalk and XLisp I spaced out on my friend's login name. He is at Cal State University Hayward, which has no news feed. He is a fanatic for smalltalk and LISP and I hope you folks out there can assist. Please no flamage, this guy is not a regular netter and he really would love some contacts. Here is what he asked me to post. ***************************************************** * e-mail responses to * * {seismo,ihnp4,pyramid}!lll-crg!csuh!jeff * * -or- * * hplabs!qantel!csuh!jeff * ***************************************************** #1 To all people,places, and things who possess some knowledge about Lsmalltalk: I am just getting into Lsmalltalk and I am interested in communicating with others who have some experience with it. I am using Smalltalk 'blue' as my map of the Lsmalltalk system; can anyone suggest a way around class-variables and methods ( is the class Smalltalk the only way?). Is there anyone who has done some interesting applications they would like to share? jeff #2 The young and struggling C.S. department of the Calif. State University of Hayward would like to get to Xlisp. If somebody out there knows were we can get it, could you please pass that information along? jeff ------------------------------ Date: 23 Sep 86 01:00:29 GMT From: zeus!stiber@locus.ucla.edu (Michael D Stiber) Subject: Kyoto Common Lisp Does anyone have experience using this Lisp, or have any information about it. I am specifically interested in comments on Ibuki Lisp, an implementation of Kyoto Common Lisp that runs on the IBM RT. Michael Stiber ARPANET: stiber@ucla-locus.arpa USENET: ...{ucbvax,ihpn4}!ucla-cs!stiber Born too late to be a yuppy -- and proud of it! ------------------------------ Date: Wed, 24 Sep 86 08:32:13 edt From: jlynch@nswc-wo.ARPA Subject: LISP Conversion I am gathering information concerning the conversion or translation of programs written in LISP to FORTRAN. Would appreciate comments from anyone who has tried to do this and the likelihood of success. Interested in both manual methods as well as conversion routines or programs. I will summarize replies for the AILIST. Thanks, Jim Lynch (jlynch@nswc-wo.arpa) ------------------------------ Date: Mon, 22 Sep 86 11:03:20 -0500 From: schwamb@mitre.ARPA Subject: Cognitive Science Schools Well, now that some folks have commented on the best AI schools in the country, could we also hear about the best Cognitive Science programs? Cog Sci has been providing a lot of fuel for thought to the AI community and I'd like to know where one might specialize in this. Thanks, Karl (schwamb@mitre) ------------------------------ Date: 18 Sep 86 13:38:33 GMT From: gilbh%cunyvm.bitnet@ucbvax.Berkeley.EDU Subject: Re: AI Grad Schools One might consider CUNY (City University of New York) too. ------------------------------ Date: Sat, 20 Sep 86 07:39:34 MDT From: halff@utah-cs.arpa (Henry M. Halff) Subject: Re: Any OPS5 in PC ? In article <8609181610.AA08808@ucbvax.Berkeley.EDU>, EDMUNDSY%northeastern.edu@RELAY.CS.NET writes: > Does anyone know whether there is any OPS5 software package availiable in PC? > I would like to know where I can find it. Thanks!!! Contact Computer*Thought 1721 West Plano Parkway Suite 125 Plano, TX 75075 214/424-3511 ctvax!mark.UUCP Disclaimer: I know people at Computer*Thought, but I don't know anything about their OPS-5. I don't know how well it works. I don't even know if I would know how to tell how well it works. ------------------------------ Date: Fri, 19 Sep 86 06:15:37 cdt From: mlw@ncsc.ARPA (Williams) Subject: OPSx for PCs For parties seeking OPS5 on PCs...an implementation of OPS/83 is being marketed by Production Systems Technologies, Inc. 642 Gettyburg Street Pittsburgh, PA 15206 (412)362-3117 I have no comparison information relating OPS5 to OPS83 other than the fact that OPS83 is compiled and is supposed to provide better performance in production on micros than is possible with OPS5. I'd be glad to see more information on the topic in this forum. Usual disclaimers... Mark L. Williams (mlw @ncsc.arpa) ------------------------------ Date: 18 Sep 86 19:21:50 GMT From: ssc-vax!bcsaic!pamp@uw-beaver.arpa (wagener) Subject: Re: Info on UNIX based AI Tools/applications (2nd req) In article <1657@ptsfa.UUCP> jeg@ptsfa.UUCP (John Girard) writes: > >This is a second request for information on Artificial Intelligence >tools and applications available in the unix environment. > > Expert System Shells > Working AI applications (academic and commercial) I can recomend at least one good comprehensive listing of tools,languages and companies; The International Directory of Artificial Intelligence Companies,2nd edition,1986,Artificial Intelligence Software S.R.L.,Via A. Mario,12/A, 45100 ROVIGO, Italy. Della Jane Hallpike,ed. Ph.(0425)27151 It mainly looks at the companies, but it does have descriptions of their products. Also look into D.A.Waterman's book, A guide to expert systems; Addison-Wesley Pub.Co.,1985. I also recomend you check out the Expert system Magazines; 1) Expert Systems - The Ineternational Journal of Knowledge Engineering;Learned Information Ltd., (This is an English Publication. It's US office address is; Learned information Co. 143 Old Marlton Pike Medfor,NJ 08055 PH.(609) 654-6266 Subscription Price: $79 2) Expert Systems User; Expert Systems User Ltd. Cromwell House, 20 Bride Lane London EC4 8DX PH.01-353 7400 Telex: 23862 Subscription Price: $210 3) IEEE Expert - Intelligent Systems and their Applications IEEE Computer Society IEEE Headquartes 345 East 47th Street New York,NY 10017 IEEE Computer Society West Coast Office 10662 Los Vaqueros Circle Los Alamitos, CA 90720 Subscription Price (IEEE Members): $12/yr 4) AI Expert AI Expert P.O.Box 10952 Palo Alto, CA 94303-0968 Subscription Price: $39/yr $69/2yr $99/3yr There are some good product description sections and articles in these (especially the British ones which are the older publications). There are quite a number of systems out there. Good luck. Pam Pincha-Wagener ------------------------------ Date: 20 Sep 86 15:44:00 edt From: Walter Hamscher Subject: queries about expert systems Date: Thu, 18 Sep 1986 17:10 EDT From: LIN@XX.LCS.MIT.EDU 1. Production systems are the implementation of many expert systems. In what other forms are "expert systems" implemented? [I use the term "expert system" to describe the codification of any process that people use to reason, plan, or make decisions as a set of computer rules, involving a detailed description of the precise thought processes used. If you have a better description, please share it.] ``Expert System'' denotes a level of performance, not a technology. The particularly important aspirations are generality and robustness. Every program strives for some degree of generality and robustness, of course, but calling a program an expert system means it's supposed to be able to do the right thing even in situations that haven't been explicitly anticipated, where ``the right thing'' might just be to gracefully say ``I dunno'' when, indeed, the program doesn't have the knowledge needed to solve the problem posed. Production systems, or, more accurately, programs that work by running a simple interpreter over a body of knowledge represented as IF-THEN rules, ease the construction of simple expert systems because it's possible to encode the knowledge without having to commit to a particular order or context of using that knowledge. The interpreter determines what rule to apply next at runtime, and so long as you don't include contradictory rules or assume a particular order of application, such systems are easy to construct and work pretty well, i.e. can be general (solve a wide variety of problem instances) and robust (degrade gracefully by saying ``i dunno'' (no rules, or only very general rules apply) in unusual situations, rather than trapping out with an error). That may not have seemed like an answer to question #1, so let me return to it explicitly. Production systems are not the only technology for building expert systems, but pattern-directed invocation is a theme common to all expert systems, whatever technology is used. Let me explain. Another popular technology for expert systems (in the medical domain, especially) might be called Frames and Demons. Facts are organized in a specialization hierarchy, and attached to each fact may be a bunch of procedures (demons) that are run when the fact is asserted, or denied, when the program needs to figure out whether the fact is true or not, etc. Running a demon may trigger other demons, or add new facts, or new demons, and so the system grinds away. The underlying principle is the same as in production systems: there is a large body of domain specific knowledge, plus a simple interpreter that makes no initial commitment to the order or context in which the facts are going to be used. The name of the game is pattern-directed invocation: the next action to take is selected from among the ``rules'' or ``methods'' or ``demons'' that are relevant to the current situation. This characteristic is not unique to expert systems, but (I think) every program that has ever been called an expert system has this characteristic in common, and moreover that it was central to its behavior. 2. A production system is in essence a set of rules that state that "IF X occurs, THEN take action Y." System designers must anticipate the set of "X" that can occur. What if something happens that is not anticipated in the specified set of "X"? I assert that the most common result in such cases is that nothing happens. Am I right, wrong, or off the map? In most implementations of production systems, if the current situation is such that no rules match it, nothing happens (maybe the program prints out the atom 'DONE :-). If the system is working in a goal-directed fashion (e.g. it's trying to find out under what circumstances it can take action Y (action Y might be "conclude that Z has occurred")) and there aren't any rules that tell it anything about Y, again, nothing happens: it can't conclude Z. In practice, there are always very general rules that apply when nothing else does. Being general, they're probably not very helpful: "IF () THEN SAY Take-Two-Aspirin-And-Call-Me-In-The-Morning." The same applies to any brand of pattern-directed invocation. However, it's getting on the hairy edge of matters to say "System designers must anticipate the set of X that can occur." The reason is that productions (methods, demons) are supposed to be modular; independent of other productions; typically written to trigger on only a handful of the possibly thousands of features of the current situation. So in fact I don't need to anticipate all the situations that occur, but rather ``just'' figure out all the relevant features of the space of situations, and then write rules that deal with certain combinations of those features. It's like a grammar: I don't have to anticipate every valid sentence, except in the sense that I need to figure out what all the word categories are and what local combinations of words are legal. Now, to hone your observation a bit, I suggest focusing on the notion of ``figuring out all the relevant features of the space of situations.'' That's what's difficult. Experts (including carbon-based ones) make mistakes when they ignore (or are unaware of) features of the situation that modify or overrule the conclusions made from other features. The fundamental problem in building an expert system that deals with the real world is not entirely in cramming enough of the right rules into it (although that's hard), it's encoding all the exceptions, or, more to the point, remembering to include in the program's model of the world all the features that might be relevant to producing exceptions. End of overly long flame. Walter Hamscher P.S. I am not an AI guru, rather, a mere neophyte disciple of the bona fide gurus on my thesis committee. ------------------------------ Date: Tue Sep 23 11:33:13 GMT+1:00 1986 From: mcvax!lasso!ralph@seismo.CSS.GOV (Ralph P. Sobek) Subject: Re: queries about expert systems (Vol 4, no. 187) Herb, >1. Production systems are the implementation of many expert systems. >In what other forms are "expert systems" implemented? I recommend the book "A Guide to Expert Systems," by Donald Waterman. It describes many expert systems, which fall more or less into your definition, and in what they are implemented. Production Systems (PSs) can basically be divided into forward-chaining (R1/XCON) and backward-chaining (EMYCIN); mixed systems which do both exist. Other representations include frame-based (SRL), semantic nets (KAS), object- oriented, and logic-based. The representation used often depends on what is available in the underlying Expert System Tool. Tools now exist which provide an intergrated package of representation structures for the expert system builder to use, e.g., KEE and LOOPS. Expert systems are also written in standard procedural languages such as Lisp, C, Pascal, and even Fortran. >2. A production system is in essence a set of rules that state that >"IF X occurs, THEN take action Y." System designers must anticipate >the set of "X" that can occur. What if something happens that is not >anticipated in the specified set of "X"? I assert that the most >common result in such cases is that nothing happens. In both forward-chaining and backward-chaining PSs nothing happens. If the PS produces "X" then we can verify if "X" is never used. In the general case, if "X" comes from some arbitrary source there is no guarantee that the PS (or any other system) will even see the datum. Ralph P. Sobek UUCP: mcvax!inria!lasso!ralph@SEISMO.CSS.GOV ARPA: sobek@ucbernie.Berkeley.EDU (automatic forwarding) BITNET: SOBEK@FRMOP11 ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:38:17 1986 Date: Tue, 30 Sep 86 20:38:06 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #196 Status: R AIList Digest Thursday, 25 Sep 1986 Volume 4 : Issue 196 Today's Topics: Linguistics - NL Generation, Logic - TMS, DDB and Infinite Loops, AI Tools - Turbo Prolog & Xerox vs Symbolics, Philosophy - Associations & Intelligent Machines ---------------------------------------------------------------------- Date: Mon, 22 Sep 86 10:31:23 EDT From: "William J. Rapaport" Reply-to: rapaport@sunybcs.UUCP (William J. Rapaport) Subject: followup on NL generation In article lb0q@ANDREW.CMU.EDU (Leslie Burkholder) writes: >Has work been done on the problem of generating relatively idiomatic English >from sentences written in a language for first-order predicate logic? >Any pointers would be appreciated. > >Leslie Burkholder >lb0q@andrew.cmu.edu We do some work on NL generation from SNePS, which can easily be translated into pred. logic. See: Shapiro, Stuart C. (1982), ``Generalized Augmented Transition Network Grammars For Generation From Semantic Networks,'' American Journal of Computational Linguistics 8: 12-25. William J. Rapaport Assistant Professor Dept. of Computer Science, SUNY Buffalo, Buffalo, NY 14260 (716) 636-3193, 3180 uucp: ..!{allegra,decvax,watmath,rocksanne}!sunybcs!rapaport csnet: rapaport@buffalo.csnet bitnet: rapaport@sunybcs.bitnet ------------------------------ Date: 20 Sep 86 15:41:26 edt From: Walter Hamscher Subject: TMS, DDB and infinite loops Date: Mon, 08 Sep 86 16:48:15 -0800 From: Don Rose Does anyone know whether the standard algorithms for belief revision (e.g. dependency-directed backtracking in TMS-like systems) are guaranteed to halt? That is, is it possible for certain belief networks to be arranged such that no set of mutually consistent beliefs can be found (without outside influence)? I think these are two different questions. The answer to the second question depends less on the algorithm than on whether the underlying logic is two-valued or three-valued. The answer to the first question is that halting is only a problem when the logic is two-valued and the space of beliefs isn't fixed during belief revision [Satisifiability in the propositional calculus is decidable (though NP-complete)]. Doyle's TMS goes into infinite loops. McAllester's won't. deKleer's ATMS won't loop either, but that's because it finds all the consistent labelings, and there just might not be any. Etc, etc; depends on what you consider ``standard.'' Walter Hamscher ------------------------------ Date: Sat, 20 Sep 86 15:02 PDT From: dekleer.pa@Xerox.COM Subject: TMS, DDB and infinite loops question. Does anyone know whether the standard algorithms for belief revision (e.g. dependency-directed backtracking in TMS-like systems) are guaranteed to halt? It depends on what you consider the standard algorithms and what do you consider a guarantee? Typically a Doyle-style (NMTMS) comes in two versions, (1) guaranteed to halt, and, (2) guaranteed to halt if there are no "odd loops". Version (2) is always more efficient and is commonly used. The McAllester-style (LTMS) or my style (ATMS) always halt. I don't know if anyone has actually proved these results. That is, is it possible for certain belief networks to be arranged such that no set of mutually consistent beliefs can be found (without outside influence)? Sure, its called a contradiction. However, the issue of what to do about odd loops remains somewhat unresolved. By odd loop I mean a node which depends on its own disbelief an odd number of times, the most trivial example being give A a non-monotonic justification with an empty inlist and an outlist of (A). ------------------------------ Date: Tue 23 Sep 86 14:39:47-CDT From: Larry Van Sickle Reply-to: CS.VANSICKLE@R20.UTEXAS.EDU Subject: Money back on Turbo Prolog Borland will refund the purchase price of Turbo Prolog for sixty days after purchase. The person I talked to at Borland was courteous, did not argue, just said to send the receipt and software. Larry Van Sickle U of Texas at Austin cs.vansickle@r20.utexas.edu 512-471-9589 ------------------------------ Date: Tue 23 Sep 86 13:54:29-PDT From: Ken Laws Reply-to: AIList-Request@SRI-AI.ARPA Subject: Turbo Prolog For another review of Turbo Prolog see the premier issue of AI Expert. Darryl Rubin discusses several weaknesses relative to Clocksin-and-Mellish prologs, but is enthusiastic about the package for users who have no experience with (i.e., preconceptions from) other prologs. The Turbo version is very fast, quite compact, well documented, comes with a lengthy library of example programs, and interfaces to a sophisticated window system and other special tools. It could be an excellent system for database retrieval and other straightforward tasks. His chief reservation was about the "subroutine call" syntax that requires all legal arities and argument types to be predeclared and does not permit use of comma as a reduction operator. -- Ken Laws ------------------------------ Date: 19 Sep 86 14:27:15 GMT From: sdcrdcf!darrelj@hplabs.hp.com (Darrel VanBuer) Subject: Re: Dandelion vs Symbolics A slight echo on the Interlisp file package (partly response to earlier note on problems using MAKEFILE, and losing a bunch of user-entered properties. Rule 1. Users never call MAKEFILE (in 9 years of Interlisp hacking, I've probably called it half a dozen times). So how do you make files? I mainly use two functions: CLEANUP() or CLEANUP(file1 file2 ...) Former does all files containing modifications, latter only named files. The first thing CLEANUP does is call UPDATEFILES, which is also called by: FILES?() Reports the files which need action to have up to date source, compiled and hardcopies, also calls UPDATEFILES, which will engage you in a dialog asking you the location of every "new" object. Most of the ways to define or modify objects are "noticed" by the file package (e.g. the structure editor [DF, EF, DV ...], SETQ, PUTPROP, etc which YOU type at top level). When an object is noticed as modified, either the file(s) containing it are marked as needing a remake, or it gets noted as something to ask you about later. You can write functions which play the game by calling MARKASCHANGED as appropriate. Two global variables interact with details of the process: RECOMPILEDEFAULT usually EXPRS or CHANGES. I prefer the former, but CHANGES has been the default in Interlisp-D because EXPRS didn't work before Intermezzo. CLEANUPOPTIONS My setting is usually (RC STF LIST) which means as part of cleanup, recompile, with compiler flags STF (F means forget source from in core, filepkg will automagically retrieve it if you edit, etc), and make a new hardcopy LISTing. For real fun with filepkg and integration with other tools, try MASTERSCOPE(ANALYZE ALL ON file1) MASTERSCOPE(EDIT WHERE ANY CALLS FOO) CLEANUP() -- Darrel J. Van Buer, PhD System Development Corp. 2525 Colorado Ave Santa Monica, CA 90406 (213)820-4111 x5449 ...{allegra,burdvax,cbosgd,hplabs,ihnp4,orstcs,sdcsvax,ucla-cs,akgua} !sdcrdcf!darrelj VANBUER@USC-ECL.ARPA ------------------------------ Date: Sat, 20 Sep 86 10:23:18 PDT From: larus@kim.Berkeley.EDU (James Larus) Subject: Symbolics v. Xerox OK, here are my comments on the Great Symbolics-Xerox debate. [As background, I was an experienced Lisp programmer and emacs user before trying a Symbolics.] I think that the user interface on the Symbolics is one of the poorest pieces of software that I have ever had the misfortune of using. Despite having a bit-mapped display, Symbolics forces you to use a one-window on the screen at a time paradigm. Not only are the default windows too large, but some of them (e.g. the document examiner) take over the whole screen (didn't anyone at Symbolics think that someone might want to make use of the documentation without taking notes on paper?). Resizing the windows (a painful process involving a half-dozen mouse-clicks) results in unreadable messages and lost information since the windows don't scroll (to be fixed in Genera 7). I cannot understand how this interface was designed (was it?) or why people swear by it (instead of at it). The rest of the system is better. Their Common Lisp is pretty solid and avoids some subtle bugs in other implementations. Their debugger is pretty weak. I can't understand why a debugger that shows the machine's bytecodes (which aren't even documented for the 3600 series!) is considered acceptable in a Lisp environment. Even C has symbolic debuggers these days! Their machine coexists pretty well with other types of systems on an internet. Their local filesystem is impressively slow. The documentation is pretty bad, but is getting better. It reminds me of the earlier days of Unix, where most of the important stuff wasn't written down. If you had an office next to a Unix guru, you probably thought Unix was great. If you just got a tape from Bell, then you probably thought Unix sucked. There appears to be a large amount of information about the Symbolics that is not written down and is common knowledge at places like MIT that successfully use the machines. (Perhaps Symbolics should ship a MIT graduate with their machines.) We have had a lot of difficulty setting up our machines. Symbolics has not been very helpful at all. /Jim ------------------------------ Date: Tue Sep 23 12:31:35 GMT+1:00 1986 From: mcvax!lasso!ralph@seismo.CSS.GOV (Ralph P. Sobek) Subject: Re: Xerox 11xx vs. Symbolics 36xx vs. ... I enjoyed all the discussion on the pluses and minuses of these and other lisp machines. I, myself, am an Interlisp user. Those who know a particular system well will prefer it over another. All these lisp systems are quite complex and require a long time, a year or so, before one achieves proficiency. And as any language, human or otherwise, one's perception of one's environment depends upon the tools/semantics that the language provides. I have always found Interlisp much more homogeneous than Zetalisp. The packages are structured so as to interface easily. I find the written documentation also much more structured, and smaller, than the number of volumes that come with a Symbolics. Maybe, Symbolics users only use the online documentation and thus avoid the pain of trying to find something in the written documentation. The last time I tried to find something in the Symbolics manuals -- I gave up, frustrated! :-) Interesting will be the future generation of lisp machines, after Common Lisp. Ralph P. Sobek UUCP: mcvax!inria!lasso!ralph@SEISMO.CSS.GOV ARPA: sobek@ucbernie.Berkeley.EDU (automatic forwarding) BITNET: SOBEK@FRMOP11 ------------------------------ Date: 22 Sep 86 12:28:00 MST From: fritts@afotec Reply-to: Subject: Associations -- Comment on AIList Digest V4 #186 The remark has been made on AIList, I think, and elsewhere that computers do not "think" at all like people do. Problems are formally stated and stepped through sequentially to reach a solution. Humans find this very difficult to do. Instead, we seem to think in a series of observations and associations. Our observations are provided by our senses, but how these are associated with stored memory of other observations is seemingly the key to how humans "think". I think that this process of sensory observation and association runs more or less continuously and we are not conciously aware of much of it. What I'd like to know is how the decision is made to associate one observation with another; what rules of association are made and are they highly individualized or is there a more general pattern. How is it that we acquire large bodies of apparently diverse observations under simple labels and then make complex decisions using these simple labels rather than stepping laboriously through a logical sequence to achieve the same end? There must be some logic to our associative process or we could not be discussing this subject at all. Steve Fritts FRITTS@AFOTEC ------------------------------ Date: 22 Sep 86 09:01:50 PDT (Monday) From: "charles_kalish.EdServices"@Xerox.COM Subject: Re: intelligent machines In his message, Peter Pirron sets out what he believes to be necessary attributes of a machine that would deserved to be called intelligent >From my experience, I think that his intuitions about what it would take for for a machine to be intelligent are, by and large, pretty widely shared and as far as I'm concerned, pretty accurate. Where we differ, though, is in how these intuitions apply to designing and demonstrating machine intelligence. Pirron writes: "There is the phenomenon of intentionality amd motivation in man that finds no direct correspondent phenomenon in the computer." I think it's true that we wouldn't call anything intelligent we didn't believe had intentions (after all intelligent is an intentional ascription). But I think that Dennet (see "Brainstorms") is right in that intentions are something we ascribe to systems and not something that is built in or a part of that system. The problem then becomes justifying the use of intentional descriptions for a machine; i.e. how can I justify my claim that "the computer wants to take the opponent's queen" when the skeptic responds that all that is happening is that the X procedure has returned a value which causes the Y procedure to move piece A to board position Q? I think the crucial issue in this question is how much (or whether) the computer understands. The problem with systems now is that it is too easy to say that the computer doesn't understand anything, it's just manipulating markers. That is that any understanding is just conventional -- we pretend that variable A means the Red Queen, but it only means that to us (observers) not to the computer. How then could we ever get something to mean anything to a computer? Some people (I'm thinking of Searle) would say you can't, computers can't have semantics for the symbols they process. I found this issue in Pirron's message where he says: "Real "understanding" of natural language however needs not only linguistic competence but also sensory processing and recognition abilities (visual, acoustical). Language normally refers to objects which we first experience by sensory input and then name it." The idea is that you want to ground the computer's use of symbols in some non-symbolic experience. Unfortunately, the solution proposed by Pirron: "The constructivistic theory of human learning of language by Paul Lorenzen und O. Schwemmer (Erlanger Schule) assumes a "demonstration act" (Zeigehandlung) constituting a fundamental element of man (child) learning language. Without this empirical fundament of language you will never leave the hermeneutic circle, which drove former philosphers into despair." ( having not read these people, I presume the mean something like pointing at a rabbit and saying "rabbit") has been demonstrated by Quine (see "Word and Object") to keep you well within the circle. But these arguments are about people, not computers and we do (at least feel) that the symbols we use and communicate with are rooted in non-symbolic something. I can see two directions from this. One is looking for pre-symbolic, biological constraints; Something like Rosch's theory of basic levels of conceptualization. Biologically relevant, innate concepts, like mother, food, emotions, etc. would provide the grounding for complex concepts. Unfortunately for a computer, it doesn't have an evolutionary history which would generate innate concepts-- everything it's got is symbolic. We'd have to say that no matter how good a computer got it wouldn't really understand. The other point is that maybe we do have to stay within this symbolic "prison-house" after all event the biological concepts are still represented, not actual (no food in the brain just neuron firings). The thing here is that, even though you could look into a person's brain and, say, pick out the neural representation of a horse, to the person with the open skull that's not a representation, it constitutes a horse, it is a horse (from the point of view of the neural sytem). And that's what's different about people and computers. We credit people with a point of view and from that point of view, the symbols used in processing are not symbolic at all, but real. Why do people have a point of view and not computers? Computers can make reports of their internal states probably better than we. I think that Nagel has hit it on the head (in "What is it like to be a Bat" I saw this article in "The Minds I") with his notion of "it is (or is not) like something to be that thing." So it is like something to be a person and presumably is not like something to be a computer. For a machine to be intelligent and truly understand it must be like something to be that machine. Only then can we credit that machine with a point of view and stop looking at the symbols it uses as "mere" symbols. Those symbols will have content from the machine's point of view. Now, how does it get to be like something to be a machine? I don't know but I know it has a lot more to do with the Turing test than what kind of memory orgainization or search algorithms the machine uses. Sorry if this is incoherent, but it's not a paper so I'm not going to proof it. I'd also like to comment on the claim that: " I would claim, that the conviction mentioned above {that machines can't equal humans} however philosphical or sophisticated it may be justified, is only the "RATIONALIZATION".. of understandable but irrational and normally unconscious existential fears and need of human beings" but this message is too long anyway. Suffice it too say that one can find a nasty Freudian interpretation of any point. I'd appreciate hearing any comments on the above ramblings. -Chuck ARPA: chuck.edservices@Xerox.COM ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:38:39 1986 Date: Tue, 30 Sep 86 20:38:26 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #197 Status: R AIList Digest Thursday, 25 Sep 1986 Volume 4 : Issue 197 Today's Topics: AI Tools - University of Rochester HORNE System & Parallel Inference System at Maryland, Conferences - Upcoming Conference Programs (FJCC, COMPSAC, OIS, Info. and Software Sciences, Chautaqua) ---------------------------------------------------------------------- Date: Thu, 11 Sep 86 13:30 EDT From: Brad Miller Subject: New University of Rochester HORNE system available The University of Rochester HORNE reasoning system has just been rereleased in common-lisp form, currently running on a symbolics (though any common-lisp system should be able to run it with minor porting). Features: Horne Clause resolution prover (similar to PROLOG) with typed unification and specialized reasoner for equalities (e.g. A and B can be asserted to be equal, and so will unify). Equalities can be asserted between any ground forms including functions with ground terms. A forward chaining proof mechanism, and an interface between this system and arbitrary common-lisp forms are also provided. As part of the same release we are providing REP, a frame-like knowledge representation system built on top of the theorem prover, which uses sturctured types to represent sets of objects. A structured type may have relations (or "roles") between its set of objects and other sets. Arbitrary instances of an object may be asserted to be equal to another instance which will utelize the underlying HORNE equality mechanisms. HORNE is the product of several years of R&D in the Natural Language Understanding and Knowledge Representation projects supervised by Prof. James Allen at the University of Rochester, and forms the basis for much of our current implementation work. A tutorial introduction and manual, TR 126 "The HORNE reasoning system in Common-Lisp" by Allen and Miller is available for $2.50 from the following address: Ms. Peg Meeker Technical Reports Administrator Department of Computer Science 617 Hylan Building University of Rochester River Campus Rochester, NY 14627 In addition a DC300XL cartridge tape in Symbolics distribution format, or Symbolics carry-tape format (also suitable for TI Explorers), or a 1/2" 1600bpi reel in 4.2BSD TAR format (other formats are not available) is available from the above address for a charge of $100.00 which will include one copy of the TR. This charge is made to defray the cost of the tape, postage, and handling. The software itself is in the public domain. Larger contributions are, of course, welcome. Please specify which format tape you wish to receive. By default, we will send the Symbolics distribution format. All checks should be made payable to "University of Rochester, Computer Science Department". POs from other Universities are also acceptable. Refunds for any reason are not available. DISCLAIMER: The software is supplied "as-is" without any implied warrenties of merchantability or fitness for a particular purpose. We are not responsible for any consequential damages as the result of using this software. We are happy to accept bug reports, but promise to fix nothing. Updates are not included; future releases (if any) will probably be made available under a similar arrangement to this one, but need not be. In other words, what you get is what you get. Brad Miller Computer Science Department University of Rochester miller@rochester.arpa miller@ur-acorn.arpa ------------------------------ Date: Thu, 11 Sep 86 17:08:33 EDT From: Jack Minker Subject: Parallel Inference System at Maryland [Excerpted from the Prolog digest by Laws@SRI-STRIPE.] AI and Database Research Laboratory at the University of Maryland Jack Minker - Director The AI and Database Research Laboratory at the Univer- sity of Maryland is pleased to announce that a parallel logic programming system (PRISM) is now operational on the McMOB multiprocessosor. The system uses up to sixteen pro- cessors to exploit medium grained parallelism in logic pro- grams. The underlying ideas behind PRISM appeared in [Eis- inger et. al., 1982] and [Kasif et. al., 1983]. [...] If you would like further information on PRISM, please contact MINKER@MARYLAND or MADHUR@MARYLAND. We would also be very interested in hearing from people who may have prob- lems we could run on PRISM. References: 1. Eisinger, N., Kasif, S., and Minker, J., "Logic Pro- gramming: A Parallel Approach", in Proceedings of the First International Logic Programming Conference, Mar- seilles, France, 1982. 2. Kasif, S., Kohli, M., and Minker, J., "PRISM - A Paral- lel Inference System for Problem Solving", in IJCAI-83, Karlsruhe, Germany, 1983. 3. Rieger, C., Bane, j., and Trigg, R., "ZMOB: A Highly Parallel Multiprocessor", University of Maryland, TR- 911, May 1980 ------------------------------ Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: ****** AI AT UPCOMING CONFERENCES ****** AI papers at November 2-6, 1986 FJCC, Dallas Texas Professional Education Program Items John D. McGregor, Building Expert Systems Workshop Loiss Boggess and Julia Hodges, Knowledge-Based-Based Expert Systems Benjamin Wah,Architectures for AI Applications Michael Lebowitz, Natural Language Processing Michael Lebowitz, Machine Learning Paul Bamberg, Speech Recognition:From Isolated Digits to Natural Language Dissertation John Kender and Takeo Kanade, Computer Vision from an AI Perspective Douglas DeGroot, Prolog and Knowledge INfo Processing Harland H. Black, AI Programming and Environments Paper Sessions AI-1, November 4, 1:30 PM to 3:30 PM Panel Session on "Design Issues and Practice in AI Programming" AI-2 Session 1, November 5, 10:00 am to noon, Computer Vision Generic Surface Interpretation Inference Rules and Quasi-Invariants Thomas Binford, Stanford U. An Overview of Computation of Structure and Motion From Images J. K. Aggarwal, University of Texas at Austin Industrial World Vision Robert Haralick, Machine Vision International AI-2 Session 2, November 5, 1:30 PM-3:30PM Survey of Image Quality Measurements I. Abdou and N. Dusaussoy, University of Delaware A Spatial Knowledge Structure for Image Information Systems Using Symbolic Projects S. K. Chang, U. of Pittsburgh, E. Jungert, FFV Elektronic A. B. Document Image Understanding S. N. Srihari, SUNY at Buffalo AI-3 Session 1, November 5 3:45 PM- 5:15 PM, Robotics Living in a Dynamic World Russell A. Andersson, AT&T Bell Labs Error Modeling in Stereo Navigation L. Matthies and S. A. Shafer, Carnegie Mellon U CMU Sidewalk Navigation System Y. Goto, et. al. Carnegie Mellon U. AI-3 Session 2, November 6 10AM - noon Automatic Gasp Planning: An Operation Space Approach M. T. Mason R. C. Brost Carnegie Mellon U. Planning Stable Grasps for Multi-fingered Hands V. Nguyen, MIT Off-line Planning for On-line Object Localization T. Lozano-perez, W.E. Grimson, MIT AI-3 Session 3, Novmeber 6, 1:30 PM - 3:30pm AMLX: A Manufacturing Language/Extended L. Nackman, et al. IBM t. J. Watson Research Center SATYR and the NYMPH: SoftwareDesign in a Multiprocessor for Control Systems J. B. Chen et. al. Stanford University The Meglos User Interface R. Gaglianello and H. Katseff, AT&T Bell Laboratories A Robot Force and Motion Server R. Paul and H. Zhang, University of Pennsylvania AI4, Session 1 November 5, 1:30pm - 3:30 pm, Rule Based Systems The AI-ADA Interface Dr. Jorge Diaz-Herrera, George Mason University The AI-LiISP Environment Dr. Harry Tennant, Texas Instruments The AI PROLOG Environment: SIMPOS- Sequential Inference Machine Programming Drs. H. Ishibashi, T. Chikayama, H. Sato, M. Sato and S. Uchida, ICOT Research Center Software Engineering for Rule-Based systems R. J. K. Jacob and J. N. Froscher, Naval Research Laboratory Session 2: Knowledge Engineering pannel, November 5, 3:45 PM - 5:15 PM Dr. Richard Wexelblat, Philips Lab, Chair Dr. Paul Benjamin, Philips Laboratories Dr. Christina Jette, Schlumberger Well Services Dr. STeve Pollit, Digital Equipment Session 3:, November 6, 1:30PM - 3:30 PM "An Organizational Frameworks for Building Adapative Artificial Intelligence Systems: T. Blaxton and B. Kushner, BDM Corporation "An Object/Task Modelling Approach" Q. Chen, Beijing Research Institute of Surveying and Mapping "A Plant INtelligent Supervisory Control Expert System: M. Ali and E. Washington, Unviersity of Tennessee "Knowledge-Based Layout Design System for Industrial Plants" K. Yoshida, et. al., Hitachi Ltd. Session 4: Prolog and Frame based Methods, November 6, 3:45 pm to 5:15 pm "A Logic-Programming Approach to Frame-Based Language Design" H. H. Chen, I. P. Lin and C. P. Wu, National Taiwan University "Interfacing Prolog to Pascal" K. Magel, North Dakota State University "Knowledge-Based Optimization in Prolog Compiler" N. Tamura, Japan Science Institute, IBM Japan Natural Language Processing, Session 1, Nov. 4 10AM - noon "Communication with Expert Systems" Kathleen R. McKeown, Columbia University "Language Analysis in Not-So-Limited Domains" Dr. Paul S. Jacobs, General Electric, R&D "Providing Expert Systems with INtegrated Natural Language and Graphical Interfaces" Dr. Philip J. Hayes, Carnegie Group Inc. "Pragmatic Processes in a Portable NL System" Dr. Paul Martin, SRI_AI Center Session 2:Nov 4 1:30-3:30pm "Uses of Structured Knowledge Representation Systems in Natural Language Processing" N. Sondheimer, University of Southern California "Unifying Lexical, Syntactic and Semantic Text Processing" K. Eiselt, University of California at Irvine "Robustness in Natural Language Interfaces" R. Cullingford, Georgia Tech "Connectionist Approaches to Natural Language Processing" G. Cottrell, UC of San Diego Panel: Problems and Prospects of NLP NOvember 4, 3:45pm - 5:15pm Chair: Dr. Philip J. Hayes Gene Charniak, Brown University, Dave Waltz Thinking Machines Robert Wilensky, UC at Berkeley, Gary Hendrix, Symantec, Jerry Hobbs, SRI "Parallel Processing for AI" Tuesday November 4 10am - 12noon "Parallel Prodcessing of a Knowledge-Based Vision System" D. I. Moldovan and C. I. Wu, USC "A Fault Tolerant, Bit-Parallel, Cellular Array Processor" S. Morton, ITt-Advanced Technology Center "Implementation of Parallel Prolog onTree Machines" M. Imai, Toyohashi University of Technology "Optimal Granularity of Parallel Evaluation of AND-Trees" G. J. Li and B. W. Wah, University of Illinois at Urbana (some of the following sessions contain non-AI papers that are not listed) Session 2: New Directions in Optical Computing" November 4 1:30pm - 3:30 pm "Optical Symbolic Computing" Dr. John Neff, DARPA/DSO andB. Kushner, BDM Co. VLSI Design and Test: Theory and Practice, Nov 4 10AM - 12 noon A Knowledge-Based TDM Selection System M. E. Breuer and X. Zhu, USC Expert Systems for Design and Test Thursday, November 6, 10AM - 12 noon DEFT, A Design for Testability Expert System J. A. B. Fortes and M. A. Samad Experiences in Prolog DFT Rule Checking G. Cabodi, P. Camurati and P. Prinetto, Politecnico di Torino Object-Oriented Software, Tuesday, November 4 1:30pm - 3:30 pm "Some Problems with Is-A: Why Properties are Objects" Prof. Stan Zdonik, Brown University Computer Chess Techniques "Phased State Space Search" T. A. Marsland, University of Alberta and N. Srimani, Southern Illinois U. "A MultiprocessorChess Program" J. Schaeffer, University of Alberta Panel Discussion Tony Marsland, U. of Alberta, Hans Berliner, CMU, Ken Thompson, AT&T Bell Labs Prof. Monroe Newborn, McGill University, David Levy, IntelligentSoftware, Prof. Robert Hyatt,U. of Southern Mississippi Searching, Nov 6, 10AM - 12 noon "Combining Symmetry and Searching", L. Finkelstein, et al. Northeastern University Fifth Generation Computers I: Language Arch, Nov 5, 10AM - 12 noon Knowledge-Based Expert System for Hardware Logic Design T. Mano et. al. Fujitsu Research Activities on Natural Language Processing of the FGCS Project H. Miyhoshi, et al., ICOT ARGOS/V: A System for Verification of Prolog Programs H. Fujita, et al., Mitsubishi Electric Session 4: "Supercomputing Systems" November 6 10:00am - noon The IX Supercomputer for Knowledge Based Systems T. Higuchi, et al. ETL (There are positions as volunteers available for which you get to attend the conference and get a copy of the proceedings in exchange for helping out one day. If interested call, 409-845-8981. The program is oriented towards graduate students and seniors.) __________________________________________________________________________ Compsac 86, Conference October 7-10 1986, Americana Congress Hotel, Chicago Ill Tutorial: October 6, 1986, 9AM - 5PM Doug DeGroot, Prolog and Knowledge Information Processing October 8 11:00 AM - 12:30 PM Modularized OPS-Based Expert Systems Using Unix Tools Pamela T. Surko, AT&T Bell Labs Space Shuttle Main Engine Test Analysis: A Case Study for Inductive Knowledge Based systems for Very Large Databases Djamshid Asgari, Rockwell International Kenneth L. Modesitt, California State University A Knowledge Based Software Maintenance Environment; Steven S. Yau, Sying-Syang Liu, Northwestern University October 8 2:00PM - 3:30 PM An Evaluation of Two New INference Control Methods Y. H. Chin, W. L. Peng, National Tsing Hua University, Taiwan Learning Dominance Relations inCombinatorial Search Problems Chee-Fen Yu, Benjamin Wah of University of Illinois at Urbana-Champaign, USA Fuzzy Reasoning Based on Lambda-LH Resolution Xu-Hua Liu, Carl K. Chang and Jing-Pha Tsai, University of Illinois at Chicago 4:00-5:30PM Panel Discussion on the Impact of Knowledge-Based Technology Chair: Charl K. Chang, University of Illinois at Chicago Panelists: Don McNamara GE Corporate Research, Kiyoh Nakamura, Fujitsu (Japan), Wider Yu, AT&T Bell Labs, R. C. T. Lee, National Hsing Hua Univeristy, Taiwan Thursday, October 9, 1986, 10:30 - 12:00 PM Special Purpose Computer Systems for Supporting AI Applications Minireview by Benjamin Wah, University of Illinois at Urbana-Champaign __________________________________________________________________________ ACM Conference on Office Information Systems, October 6-8 1986, Providence Rhode Island October 6, 1986 2:45 - 4:5PM Adaptive Interface Design: A Symmetric Model and a Knowledge-Based Implementation Sherman W. Tyler, Siegfried Treu, University of Pittsburgh Automating Review of Forms for International Trade Transactions: A Natural Language Processing Approach V. Dhar, P. Ranganathan October 8, 1986 9:10:15AM Panel on "AI in the Office", Chair Gerald Barber October 8, 1986 10:30 - 12:00 AM Organizational Analysis: Organizational Ecology Modelling Due Process in the Workplace Elihu M. Gerson, Susan L. Star, Tremont Research Institute An Empirical Study of the INtegration of Computing into Routine Work Les Gasser, University of Southern California Offices are Open Systems Carl Hewitt, MIT Artificial Intelligence Lab October 8, 1986 1:00 - 2:30PM Handling Shared Resources in a Temporal Data Base Management System Thomas L. Dean, Brown University Language Constructs for Programming by Example Robert V. Rubin, Brown University Providing Intelligent Assistance in Distributed Office Environments Sergei Nirenburg, Victor Lessor Colgate University/ University of Massachusett __________________________________________________________________________ Fourth Symposium on Empirical Foundations of Information and Software Sciences October 22-24 Atlanta Georgia October 22, 1:30-3:15 PM Expert Systems for Knowledge Engineering: Modesof Development Glynn Harmon, University of Texas, Austin October 23, 10:45 - 12:30 AM Face to Machine Interaction in Natural Language: Empirical Results of Field Studies with an English and German Interface Juergen Krause, Universitaet Regensburg, F. R. Germany October 24, 9:00 - 10:30AM Evaluating Natural Language INterfaces to Expert Systems Ralph M. Weischedel BBN, Cambridge MA Couting Leaves: An Evaluation of Ada, LISP and Prolog Jagdish C. Agrawal, Embry-Riddle Aeronautical University, Daytona Beach, FL Shan Manicam, Western Carolina University, Cullowhee, NC __________________________________________________________________________ The Fourth Chautaqua, October 25-29, 1986, Coronado, California Session 9 Knowledge Based Systems 10:30-12:30PM Knowledge-based Systems Development of CNC Software, Roy Tsui, Software R&D Engineer, Pneumo Precision, INc., Allied Company Towards Resident Expertise in Systems Design Dr. Medhat Karima, CAD/CAM Consultant, Ontario CAD/CAM Center The Engineer as an Expert System Builder Dr. Richard Rosen, Vice President, Product Development, Silogic Inc. An Overview of Knowledge-Based Systems for Design and Manufacturing Dr. Larry G. Richards, Director, Master's Program, University of Virginia ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:40:13 1986 Date: Tue, 30 Sep 86 20:40:03 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #198 Status: R AIList Digest Friday, 26 Sep 1986 Volume 4 : Issue 198 Today's Topics: Correction - Learned Information Address, Queries - Computers and Writing & Prospector Shell for IBM-PC & Learning via ES Rule Refinement & Character Recognition, AI Tools - OPS5 on the PC & Turbo Prolog & Xerox vs Symbolics Storage Reclaimation, Review - Spang Robinson Summary, August 1986 ---------------------------------------------------------------------- Date: Thu, 25 Sep 86 03:49:19 EDT From: Marty Lyons Subject: Address correction for ref. in Vol 4, Issue 195 Just in case someone might have problems with USPS, Medfor should read Medford below. (Actually, mail to them should get there anyway, as long as you remember the zip, but just in case...) > AIList Digest Thursday, 25 Sep 1986 Volume 4 : Issue 195 > >Date: 18 Sep 86 19:21:50 GMT >From: ssc-vax!bcsaic!pamp@uw-beaver.arpa (wagener) >Subject: Re: Info on UNIX based AI Tools/applications (2nd req) > 1) Expert Systems - The Ineternational Journal of > Knowledge Engineering;Learned Information Ltd., > (This is an English Publication. It's US office > address is; > Learned information Co. > 143 Old Marlton Pike > Medfor,NJ 08055 *** Typo... ****** This should be Medford ------------------------------ Date: Thu, 25 Sep 86 09:59 EDT From: Hirshfield@RADC-MULTICS.ARPA Subject: Computers and Writing - A Solicitation I am soliciting contributions for a volume entitled Computers and Writing: Theory and Research to be published as part of Ablex Publishing's Writing Research Series. As the title implies, the volume will be devoted to research and theoretical investigations of the interactions of computing and writing and will focus on long- range prospects. Potential contributors include Richard Mayer, Colette Daiute, Cynthia Selfe and Jim Levin. I would be pleased to hear of any papers or any ongoing studies that relate to this exciting topic. Please respond asap by net to Hirshfield at RADC-multics, or write directly to Stuart Hirshfield, Department of Mathematics and Computer Science, Hamilton College, Clinton, NY 13323. ------------------------------ Date: 25 Sep 1986 17:48 (Thursday) From: munnari!nswitgould.oz!wray@seismo.CSS.GOV (Wray Buntine) Subject: Prospector ESs for IBM-PC OK, I've seen the recent list of IBM-PC Expert System Shells, But which PROSPECTOR-type shells have the following ability to link in external routines i.e. we have some C code that provides answers for some leaf nodes I'd be grateful for any pointers re reliability and backup as well. Wray Buntine wray@nswitgould.oz.au@seismo seismo!munnari!nswitgould.oz!wray Computing Science NSW Inst. of Tech. PO Box 123, Broadway, 2007 Australia ------------------------------ Date: 26 Sep 1986 11:08-EDT From: Hans.Tallis@ml.ri.cmu.edu Subject: Learning via ES Rule Refinement? I am working in learning by refining a given set of expert system rules. Ideally the learning cycle will involve no humans in the loop. I am familiar with Politakis's SEEK work already, but pointers to other programs would be greatly appreciated. --tallis@ml.ri.cmu.edu ------------------------------ Date: Thu, 25 Sep 86 11:10:16 edt From: philabs!micomvax!peters@tezcatlipoca.CSS.GOV Reply-to: micomva!peters@tezcatlipoca.CSS.GOV (peter srulovicz) Subject: character recognition We are starting a project that will involve a fair amount of character recognition, both typed and handwritten. If anyone out there has information about public domain software or software that can be purchased please let me hear from you. email: !philabs!micomvax!peters mail: Peter Srulovicz Philips Information Systems 600 Dr. Philips Blvd St. Laurent Quebec Canada H4M-2S9 ------------------------------ Date: 26 Sep 1986 11:13:45 EDT From: David Smith Subject: OPS5 on the PC There is an OPS5 called TOPSI available for the IBM PC from Dynamic Master Systems, Inc (404)565-0771 ------------------------------ Date: Thu, 25 Sep 86 12:09:16 GMT From: Gordon Joly Subject: Re: What's wrong with Turbo Prolog Was Clocksin and Mellish handed down on tablets of stone? An which PROLOG can claim to fulfill all the theoretical goals, eg be truly declarative? Gordon Joly. INET: joly%surrey.ac.uk@cs.ucl.ac.uk EARN: joly%uk.ac.surrey@AC.UK ------------------------------ Date: 25 Sep 1986 14:45:40 EDT (Thu) From: Dan Hoey Subject: Xerox vs Symbolics -- Reference counts vs Garbage collection In AIList Digest V4 #191, Steven J. Clark responds to the statement that ``Garbage collection is much more sophisticated on Symbolics'' with his belief that ``To my knowledge this is absolutely false. S. talks about their garbage collection more, but X's is better.'' Let me first deplore the abuse of language by which it is claimed that Xerox has a garbage collector at all. In the language of computer science, Xerox reclaims storage using a ``reference counter'' technique, rather than a ``garbage collector.'' This terminology appears in Knuth's 1973 *Art of Computer Programming* and originated in papers published in 1960. I remain undecided as to whether Xerox's misuse of the term stems from an attempt at conciseness, ignorance of standard terminology, or a conscious act of deceit. The question remains of whether Interlisp-D or Zetalisp has the more effective storage reclamation technique. I suspect the answer depends on the programmer. If we are to believe Xerox, the reference counter technique is fundamentally faster, and reclaims acceptable amounts of storage. However, it is apparent that reference counters will never reclaim circular list structure. As a frequent user of circular list structure (doubly-linked lists, anyone?), I find the lack tantamount to a failure to reclaim storage. Apparently Xerox's programmers perform their own careful deallocation of circular structures (opening the cycles before dropping the references to the structures). If I wanted to do that, I would write my programs in C. I have never understood why Xerox continues to neglect to write a garbage collector. It is not necessary to stop using reference counts, but simply to have a garbage collector available for those putatively rare occasions when they run out of memory. Dan Hoey ------------------------------ Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Spang Robinson Summary, August 1986 Spang Robinson Report Summary, August 1986, Volume 2 No. 8 23 Artificial Intelligence Application Products are out and are being used by customers. Spang Robinson tracked down 92 specific applications in 56 different companies, agencies or institutions that are being used by someone other than the developers. 24 of these are in diagnostics, 22 in manufacturing, 14 in computers, 6 in geology, 6 in chemistry, 5 in military, 4 in agriculture, 4 in medicine and 7 in "other". DEC has 20 expert systems in use with 50 under development. IBM has six in use and 64 in development. TSA Associates that there are 1000 applications fielded on microcomuters. Dataquest claims that revenues from shell products will reach 44 million in 1986, up from 22 million in 1985. The majority of this is for product training as opposed to actual price for the product. They are estimating expert systems applications to reach ten million. AIC has sold 500 copies of Intellect, a high-end natural language package and will receive 6 to 8 million dollars of revenue in 1986. Symantec's Q&A has sold 17,000 copies of Q&A, a [micro - LEFF] product with embedded natural language. There are 24 to 30 companies with viable commercial speech recognition products with market growth between 20 and 30 percent. The 1986 market will be 20 million up from 16 million. There are 100 companies in machine vision. 1985 market is estimated at 150 million dollars. General Motors bought 50 million of these products. Also, there is a discussion of estimates of how many working expert systems there are for each expert-shell product. __________________________________________________________________________ Micro Trends Teknowledge has 2500 run-time systems. Level 5 has 50 completed applications with 200 run-time systems sold. One of these systems has 3000 rules spread across nine knowledge bases for one system. Exsys has 200 applications with 2100 run-times. __________________________________________________________________________ List of commercially available expert systems Bravo: VLSI circuit design and layout (applicon) Equinox: sheet metal design (applicon) Mechanical Advantage 1000: MCAE with intelligent sketchpad (cognition) Manufacturing and Operations Management and Financial Advisor (Palladian) Expert Manufacturing Planning Systems (Tipnis, Inc.) PlanPower: financial planning system (Applied Expert System) Planman and Database; financial planning and report writer (Sterling Wentworth Corp.) Profit Tool: financial services sales aid (Prophecy Development Corp) Stock Portfolio Analysis and Futures Price Indexing (Athena Group, NY) Newspaper Layout System (Composition Systems) CEREBRAL MANAGER: manages document release process (KODAK) ICAD: production design system (ICAD, Inc.) MORE: direct marketing advisor and evaluation of mailing lists ULTRAMAX: a self-learning expert system to optimize operations (Ultramax Corp.) TRANSFORM/IMS (applications generator in COBOL (Transform Logic, Inc.) TIMM TUNER: tuning for DEC VAXs (General Research Corporation) HYPERCALC: an intelligent spreadsheet for LISP machines (Chaparral Dallas) REFINE: knowledge based software development environment (Reasoning systems, Inc.) XMP: Expert Project Manager (XSP Corporation) LEXAN: diagnostics for injection-molded plastic parts (GE) Internally developed expert systems Computers and electronics XCON,XSEL, XSITE, configures VAX orders, checks them for accuracy and plan site layout CALLISTRO: assisting in managing resources for chip designers (DEC) DAS-LOGIC assists with logic designers COMPASS analyzes maintenance records for telephone switching system and suggests maintenance actions ???? - System for design of digital circuits (Hughes) CSS: aids in planning relocation, reinstallation and rearrangement of IBM mainframes (IBM) PINE: guides people writing reports on analysis of software problems (IBM) QMF Advisor: used by customer advisors to help customers access IMS databases (IBM) Capital Assests Movements: help move capital assets quickly OCEAN: checks orders for computer systems (NCR) Diagnostic and/or preventive maintenance systems, internal use AI-Spear: tape drives (DEC) NTC: Ethernet and DECNET networks (DEC) PIES circuit fabrication line (Fairchild) Photolithographjy advisor: photolithography steps (Hewlett-Packard) DIG Voltage Tester: digital voltage sources in testing lab (Lockheed) BDS: baseband distribution system of commuications hardware (Lockheed) ACE: telephone lines (Southwest Bell) DIAG8100 DP equipment (Travelers Insurance) ????: soup cookers (Campbell Soups) Engine Cooling Advisor: engine cooling system (DELCO Products) ???? - peripherals (Hewlett-Packard) PDS: machine processes (Westinghouse) DOC: hardware and software bug analysis for Prime 750 (Prime) ???: hardware (NCR) TITAN: TI 990 Minicomputer (Radian/TI) Radar Tracking: object tracking software for radar (Arthur D. Little/Defense Contractor) ????: circuit board (Hughes) XMAN: aircraft engines (Systems Control Technology/Air Force Logistics Command) ????: circuit fault (Maritn Marietta) ????: power system diagnosis (NASA) Manufacturing or design, internal developed ????: brushes and springs for small electric motors (Delco) ISA: schedules orders for manufacturing and delivery (DEC) DISPATCHER: schedules dispatching of parts for robots (DEC) ISI: schedules manufacturing steps in job shop (Westinghouse) CELL DESIGNERS: reconfigures factories for group technologies (Arthur Anderson) WELDSELECTOR: welding engineering (Colorodo School of Mines and TI) ????: configures aircraft electrical system components (Westinghouse) CASE: electrical connector assembly (BOEING) FACTORY LAYOUT: ADL TEST FLOW DESIGN: quality test and rework sequencing (ADL for defense contractor) PTRANS: planning computer systems (DEC/CMU) PROCESS CONTROL: monitors alkylation plant (ADL) TEST FOR STORAGE SUBSYSTEM HARDWARE: IBM ???: Capacity Planning for System 38 (IBM) ??? optimization of chemical plant for EXXON ???: manage and predict weather conditions TEXACO ???: manufacturing simulation BADGER CO. ???: expert system connected to robot HERMES (Oak Ridge National Lab) ???: nuclear fuel enhancement (Westinghouse) ???: dry dock loading (General Dynamics) Medicine, internal development ????: serum protein analysis: Helena Labs PUFF: pulmonary function test interpretation: Pacific Medical Center ONCOCIN: cancer therapy manager: Stanford Oncology Clinic CORY: diagnoses invasive cardiac testas: Cedars Sinai Medimum Center TQMSTUNE: tunes tripple quadrupole mass spectrometer (Lawrence Livermore National Labs) DENDRAL: Molecular Design, Ltd. Synchem: plans chemical synthesis tests: SUNY-Stonybrook THEORISTS: polymer properties (3M) ???: organic chemical analysis (Hewlett-Packard) APPL: real time control of chemical processes related to aircraft parts (Lockheed-Georgia) Geology Internally Developed Systems SECOFOR: drill bit sticking problems (Elf-Aquitatine) GEOX: identifies earth minerals from remotely sensed hyperspectral image data (NASA) MUDMAN: diagnoses drilling mud problems (NL Industries) oNIX and DIPMETER ADVISOR: oil well logging data related systems(Schlumberger) TOGA: analyze power transformation conditions (Radian/ for Hartford Steamboiler, Inspection and Insurance Co.) Agriculture Internally Developed Systems WHEAT COUNSELOR: diasease control (ICI) POMME: apple orchard management (VA Poly inst.) PLANT/cd and PLANT/ds: soybean diseases (University of Illinois) GRAIN MARKETING ADVISOR: (PUrdue University and TI) Military AALPS: cargo planning for aircraft (US Army) RNTDS: design command and control programs for ships (Sperry) SONAR DOME TESTING: analysis of trials of sonar systems (ADL for defence contractor) NAVEX: assistant to shuttle operations (NASA) IMAGE INTERPRETATION: analyse aerial reconniassance photos (ADL for defense contractor) Other INFORMART ADVISOR: Advises shoppers on computer purchases TVX: Teaches VMS operating systems (DEC) DECGUIDE: teaches rules for design checking (Lockheed) SEMACS: monitors Securities INdustry Automation Companies Network (SIAC/Sperry) Financial Statement Analyser: Arthur Anderson __________________________________________________________________________ Neuron Data plans to have NEXPERT running on the PC/AT, and the MICRO VAX. The new system will have frames, object hierarchies and the ability to move data among concurrently running programs which will allow them to do blackboarding. __________________________________________________________________________ Paine Webber has downgraded Symbolics from "Buy" to "Attractive" due to 'market place confusion caused by Symbolics imminent transition to gate-array-based." Intellicorp got a "neutral' rating from Paine Webber due to the fact that it runs 'unaacceptably slowly' and that 'rapid expansion and redeployment of talent may strain IntelliCorp's sale forces ability to produce' __________________________________________________________________________ Symbolics prices 3620 will sell at $49,900 and 3650 will sell for $65,900. Symbolics has introduced a product to allow developers to prevent users from accidentally accessing underlying software utilities. __________________________________________________________________________ Ibuki has announced Kyoto Common Lisp. It takes 1.4MB with the kernel in C. It costs $700.00 and runs on AT&T 3B2, Integrated Solutions, Ultrix, Suns, and 4bsd __________________________________________________________________________ Integrated Inference Machines has announced SM45000 symbolic machines. It is microcodable for various languages and costs from $39,000 to $44,000. The company claims more performance than a Symbolics. __________________________________________________________________________ reviews of Wendy B. Rauch-Hindin's two volume Artificial Intelligence in Business, Science and Industry, Artificial Intelligence Enters the Marketplace by Larry Harris and Dwight Davis. and Who's Who in Artifificial Intelligence. The latter contains 399 individual biographies as well as other info. ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:40:28 1986 Date: Tue, 30 Sep 86 20:40:20 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #199 Status: R AIList Digest Friday, 26 Sep 1986 Volume 4 : Issue 199 Today's Topics: Review - Canadian Artificial Intelligence, June 1986, Philosophy - Intelligence, Consciousness, and Intensionality ---------------------------------------------------------------------- Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Canadian Artificial Intelligence, June 1986 Summary: Report of Outgoing and Ingoing Presidents Interact R&D Starts AI division Review of 1986 Canadian AI Conference at Montreal. It had 375 people registered. Best appaer was James Delgrande of Simon Fraser University. The Canadian Society for Computational Studies of Intelligence is now up to 800 from 250 two years ago. (This was prior to including people who became members upon paying non-member fees at the Canadian AI conference). Proceedings of the 1986 Conference costs $30.00 Contents Why Kids Should Learn to Program, Elliot Soloway, Yale University Generative Structure in Enumerative Learning Systems Robert C. Holte, Brunel Univeristy, R. Michael Warton, York University Detecting Analogous Learning Ken Wellsch, Marlene Junes of University of Waterloo GUMS: A General User Modeling System Tim Finin, University of Pennsylvania Dave Drager, Arity Corporation An Efficient Tableau-Based Theorem Prover Franz Oppacher, Ed Suen of Carleton University Domain Circumscription Revisited David Etherington, Universityof British Columbia Robert Mercer, University of Western Ontario A Propositional Logic for Natural Kinds James Delgrande, Simon Fraser University Fagin and Halpern on Logical Omniscienceses: A Critique with an Alternative Robert F. Hadley Simon Fraser University Representing Contextual Dependencies in Discourse Tomek Strzalkowski, Simon Fraser University A Domain-Independent Natural Language Database Interface Yawar Ali, Raymond Aubin, Barry Hall, Bell Northern Research Natural Language Report Synthesis: An Application to Marine Weather Forecasts R. Kittredge, A. Polguere of Universite de Montreal E. Goldberg Environment Canada What's in an Answer: A Theoretical Perspectiveon Deductive Questioning Answering Lenhart Schubert, L. Watanabe of University of Alberta A New Implementation for Generalized Phrase Structure Grammar Philip Harrison, Michael Maxwell Boeing Artificial Intelligence Center TRACK: Toward a Robust Natural Language INterface Sandra Carberry, University of Delaware Representation of Negative and Incomplete Information in Prolog Kowk Hung Chan, University of Western Ontario On the Logic of Representing Dependencies by Graphs, Judea Pearl of Universityof California Azaria Paz Technion, Israel Institute of Technology A proposal of Modal Logic Programming (Extended Abstract) Seiki Akama, Fujitsu ltd., Japan Classical Equality and Prolog E. W. Elcock and P. Hoddinott of University of Western Ontario Diagnosis of Non-Syntactic Programming Errors in the Scent Advisor Gordon McCalla, Richard B. Bunt, Janelle J. Harms of University of Saskatchewan Using Relative Velocity INformation to Constrain the Motion Correspondence Problem Michael Dawson and Zenon Pylyshyn, University of Western Ontario Device Representation Using Instantiation Rules and Structural Templates Mingruey R. Taie, Sargur N. Srihari, James Geller, Stuart C. Shapro of State University of New York at Buffalo Machine Translation Between Chinese and English Wanying Jin, University of Texas at Austin Interword Constraints in Visual Word Recognition Jonathan J. Hull, State University of New York at Buffalo Sensitivity to Corners inFlow Paterns Norah K. Link and STeve Zucker, McGill University Stable Surface Estimation Peter T. Sander, STeve Zucker, McGill University Measuring Motion in Dynamic Images: A Clustering Approach Amit Bandopadhay and R. Dutta, University of Rochester Determining the 3-D Motion of a Rigid Surface Patch without Correspondence, Under Perspective Projection John Aloimonos and Isidore Rigoutsos, University of Rochester Active Navigation Amit Bandopadhay, Barun Chandra and Dana H. Ballard, University of Rochester Combining Visual and Tactile Perception for Robotics J. C. Rodger and Roger A. Browse, Queens University Observation on the Role of Constraints in Problem Solving Mark Fox of Carnegie-Mellon University Rule Interaction in Expert System Knowledge Bases Stan Raatz, University of Pennsylvania George Drastal, Rutgers University Towards User specific Explanations from Expert Systems Peter van Beek and Robin Cohen, University of Waterloo DIALECT: An Expert Assistant for Informatin REtrieval Jeane-Claude Bassano, Universite de Paris-Sud Subdivision of Knowledge for Igneous Rock Identification Brian W. Otis, MIT Lincoln Lab Eugene Freuder, University of New Hampshire A Hybrid, Decidable, Logic-Based Knowledge Representation System Peter Patel-Schneider, Schlumberger Palo Alto Research The Generalized-Concept Formalism: A Frames and Logic Based Representation Model Mira Balaban, State University of New York at Albany Knowledge Modules vs Knowledge-Bases: A Structure for Representing the Granularity of Real-World Knowledge Diego Lo Giudice and Piero Scaruffi, Olivetti Artificial Intelligence Center, Italy Reasoning in a Hierarchy of Deontic Defaults Frank M. Brown, Universityof Kansas Belief Revision in SNeps Joao P. Martins Instituto Superior Tecnico, Portugal Stuart C. Shapiro, State University of New York at Buffalo GENIAL: Un Generateur d'Interface en Langue Naturelle Bertrand Pelletier et Jean Vaucher, Universite de Montreal Towards a Domain-Independent Method of Comparing Search Algorithm Run-times H. W. Davis, R. B. Polack, D. J. Golden of Wright State University Properties of Greedily Optimized Ordering Problems Rina Dechter, Avi Dechter, University of California, Los Angeles Mechanisms in ISFI: A Technical Overview (Short Form) Gary A. Cleveland TheMITRE Corp. Un Systeme Formel de Caracterisation de L'Evolution des Connaissances Eugene Chouraqui, Centre National de la Recherche Scientifique Une Experience de l'Ingenierie de la Connaissance: CODIAPSY Developpe avec HAMEX Michel Maury, A. M. Massote, Henri Betaille, J. C. Penochet et Michelle Negre of CRIME et GRIP, Montpellier, France __________________________________________________________________________ Report on University of Waterloo Research on Logic Mediated Knowledge Based Personal Information Systems They received a 3 year $450,000 grant. They will prototype Theorist, a PROLOG based system, in which they will implement a diagnostic system with natural language interface for complex system, a system to diagnose children's reading disabilities. They will also develop a new Prolog in which to write Theorist. This group has already implemented DLOG, a "logic-based knowledge representation sytem", two Prologs (one of which will be distributed by University of wAterloo's Computer System Group), designed Theorist, implemented an expert system for diagnosing reading disabilities (which will be redone in Theoritst) and designed a new architecture for Prolog, and implemented Concurrent Prolog. __________________________________________________________________________ Reviews of John Haugeland's "Artificial Intelligence: The Very Idea" "The Connection Machine" by W W. Daniel HIllis, "Models of the Visual Cortex" by David Rose and Vernon G. Dobson ------------------------------ Date: 25 Sep 86 08:12:00 EDT From: "CUGINI, JOHN" Reply-to: "CUGINI, JOHN" Subject: Intelligence and Representation This is in response to some points raised by Charles Kalish - Allow a somewhat lengthy re-quotation to set the stage: I think that Dennet (see "Brainstorms") is right in that intentions are something we ascribe to systems and not something that is built in or a part of that system. The problem then becomes justifying the use of intentional descriptions for a machine; i.e. how can I justify my claim that "the computer wants to take the opponent's queen" when the skeptic responds that all that is happening is that the X procedure has returned a value which causes the Y procedure to move piece A to board position Q?... I think the crucial issue in this question is how much (or whether) the computer understands. The problem with systems now is that it is too easy to say that the computer doesn't understand anything, it's just manipulating markers. That is that any understanding is just conventional -- we pretend that variable A means the Red Queen, but it only means that to us (observers) not to the computer. ... [Pirron's] idea is that you want to ground the computer's use of symbols in some non-symbolic experience.... One is looking for pre-symbolic, biological constraints; Something like Rosch's theory of basic levels of conceptualization. .... The other point is that maybe we do have to stay within this symbolic "prison-house" after all event the biological concepts are still represented, not actual (no food in the brain just neuron firings). The thing here is that, even though you could look into a person's brain and, say, pick out the neural representation of a horse, to the person with the open skull that's not a representation, it constitutes a horse, it is a horse (from the point of view of the neural sytem). And that's what's different about people and computers. ... These seem to me the right sorts of questions to be asking - here's a stab at a partial answer. We should start with a clear notion of "representation" - what does it mean to say that the word "rock" represents a rock, or that a picture of a rock represents a rock, or that a Lisp symbol represents a chess piece? I think Dennett would agree that X represents Y only relative to some contextual language (very broadly construed as any halfway-coherent set of correspondence rules), hopefully with the presence of an interpreter. Eg, "rock" means rock in English to English-speakers. opp-queen means opponent's queen in the mini-language set up by the chess-playing program, as understood by the author. To see the point a bit more, consider the word "rock" neatly typed out on a piece of paper in a universe in which the English language does not and never will exist. Or consider a computer running a chess-playing program (maybe against another machine, if you like) in a universe devoid of conscious entities. I would contend that such entities do not represent anything. So, roughly, representation is a 4-place relation: R(representer, represented, language interpreter) "rock" a rock English people picture of rock a rock visual similarity people, maybe some animals ... and so on. Now... what seems to me to be different about people and computers is that in the case of computers, meaning is derivative and conventional, whereas for people it seems intrinsic and natural. (Huh?) ie, Searle's point is well taken that even after we get the chess-playing program running, it is still we who must be around to impute meaning to the opp-queen Lisp symbol. And furthermore, the symbol could just as easily have been queen-of-opponent. So for the four places of the representation relation to get filled out, to ground the flying symbols, we still need people to "watch" the two machines. By contrast two humans can have a perfectly valid game of chess all by themselves, even if they're Adam and Eve. Now people certainly make use of conventional as well as natural symbol systems (like English, frinstance). But other representers in our heads (like the perception of a horse, however encoded neurally). seem to *intrinsically* represent. Ie, for the representation relation, if "my perception of a horse" is the representer, and the horse out there in the field is the represented thing, the language seems to be a "special", natural one namely the-language-of-normal- veridical-perception. (BTW, it's *not* the case, as stated in Charles's original posting that the perception simply is the horse - we are *not* different from computers with respect to the-use-of-internal-things-to-represent-external-things.) Further, it doesn't seem to make much sense at all to speak of an "interpreter". If *I* see a horse, it seems a bit schizophrenic to think of another part of myself as having to interpret that perception. In any event, note that this is self-interpretation. So people seem to be autonomous interpreters in a way that computers are not (at least not yet). In Dennett's terminology, it seems that I (and you) have the authority to adopt an intentional stance towards various things (chess-playing machines, ailist readers, etc.), *including* ourselves - certainly computers do not yet have this "authority" to designate other things, much less themselves, as intentional subjects. Please treat the above as speculation, not as some kind of air-tight argument (no danger of that anyway, right?) John Cugini ------------------------------ Date: Thu 25 Sep 86 10:24:01-PDT From: Ken Laws Subject: Emergent Consciousness Recent philosophical discussions on consciousness and intentionality have made me wonder about the analogy between Man and Bureaucracy. Imagine a large corporation. Without knowing the full internal chain of command, an external observer could still deduce many of the following characteristics. 1) The corporation is composed of hundreds of nearly identical units (known as personnel), most of whom perform material-handling or information-handling tasks. Although the tasks differ, the processing units are essentially interchangeable. 2) The "intelligence" of this system is distributed -- proper functioning of the organization requires cooperative action by many rational agents. Many tasks can be carried out by small cliques of personnel without coming to the attention of the rest of the system. Other tasks require the cooperation of all elements. 3) Despite the similarity of the personnel, some are more "central" or important than others. A reporter trying to discover what the organization is "doing" or "planning" would not be content to talk with a janitor or receptionist. Even the internal personnel recognize this, and most would pass important queries or problems to more central personnel rather than presume to discuss or set policy themselves. 4) The official corporate spokesman may be in contact with the most central elements, but is not himself central. The spokesman is only an output channel for decisions that occur much deeper or perhaps in a distributed manner. Many other personnel seem to function as inputs or effectors rather than decision makers. 5) The chief executive officer (CEO) or perhaps the chairman of the board may regard the corporation as a personal extension. This individual seems to be the most central, the "consciousness" of the organization. To paraphrase Louis XV, "I am the state." It seems, therefore, that the organization has not only a distributed intelligence but a localized consciousness. Certain processing elements and their own thought processes control the overall behavior of the bureaucracy in a special way, even though these elements (e.g., the CEO) are physiologically indistinguishable from other personnel. They are regarded as the seat of corporate consciousness by outsiders, insiders, and themselves. Consciousness is thus related to organizational function and information flow rather than to personal function and characteristics. By analogy, it is quite possible that the human brain contains a cluster of simple neural "circuits" that constitute the seat of consciousness, even though these circuits are indistinguishable in form and individual functioning from all the other circuits in the brain. This central core, because of its monitoring and control of the whole organism, has the right to consider itself the sole autonomous agent. Other portions of the brain would reject their own autonomy if they were equipped to even consider the matter. I thus regard consciousness as a natural emergent property of hierarchical systems (and perhaps of other distributed systems). There is no need to postulate a mind/body dualism or a separate soul. I can't explain how this consciousness arises, nor am I comfortable with the paradox. But I know that it does arise in any hierarchical organization of cooperating rational agents, and I suspect that it can also arise in similar organizations of nonrational agents such as neural nets or computer circuitry. -- Ken Laws ------------------------------ Date: 25 Sep 1986 1626-EDT From: Bruce Krulwich Subject: semantic knowledge howdy. i think there was a discussion on searle that i missed a month or so ago, so this may be rehash. i disagree with searle's basic conjecture that he bases all of his logic on, namely that since computers represent everything in terms of 1's and 0's they are by definition storing knowledge syntactically and not semantically. this seems wrong to me. as a simple counterexample, consider any old integer stored within a computer. it may be stored as a string of bits, but the program implicitely has the "semantic" knowledge that it is an integer. similarly, the stored activation levels and connection strengths in a connectionist model simulator (or better, in a true hardware implementation) may be stored as a bunch of munerical values, but the software (ie, the model, not the simulator) semantically "knows" what each value is just as the brain knows the meaning of activation patterns over neurons and synapses (or so goes the theory). i think the same can be said for data stored in a more conventional AI program. in response to a recent post, i don't think that there is a fundamental difference between a human's knowledge of a horse and a computers manipulation of the symbol it is using to represent it. the only differences are the inherently associative nature of the brain and the amount of knowledge stored in the brain. i think that it is these two things that give us a "feel" for what a horse is when we think of one, while most computer systems would make a small fraction of the associations and would have much less knowledge and experience to associate with. these are both computational differences, not fundamental ones. none of this is to say that we are close or getting close to a seriously "intelligent" computer system. i just don't think that there are fundamental philosophical barriers in our way. bruce krulwich arpa: krulwich@c.cs.cmu.edu bitnet: krulwich%c.cs.cmu.edu@cmccvma ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:42:46 1986 Date: Tue, 30 Sep 86 20:42:42 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #200 Status: R AIList Digest Monday, 29 Sep 1986 Volume 4 : Issue 200 Today's Topics: Seminars - Chemical Structure Generation (SU) & Fuzzy Relational Databases (SMU) & General Logic (MIT) & Generic Tasks in Knowledge-Based Reasoning (MIT), Conference - Workshop on Qualitative Physics ---------------------------------------------------------------------- Date: Mon 22 Sep 86 23:39:33-PDT From: Olivier Lichtarge Subject: Seminar - Chemical Structure Generation (SU) I will be presenting my thesis defense in biophysics Thursday September 25 in the chemistry Gazebo, starting at 2:15. Solution Structure Determination of Beta-endorphin by NMR and Validation of Protean: a Structure Generation Expert System Solution structure determination of proteins by Nuclear Magnetic Resonance involves two steps. First, the collection and interpretation of data, from which the secondary structure of a protein is characterized and a set of constraints on its tertiary structure identified. Secondly, the generation of 3-dimensional models of the protein which satisfy these constraints. This thesis presents works in both these areas: one and two-dimensional NMR techniques are applied to study the conformation of @g(b)-endorphin; and Protean, a new structure generation expert system is introduced and validated by testing its performance on myoglobin. It will be shown that @g(b)-endorphin is a random coil in water. In a 60% methanol and 40% water mixed solvent the following changes take place: an @g(a)-helix is induced between residues 14 and 27, and a salt bridge forms between Lysine28 and Glutamate31, however, there still exists no strong evidence for the presence of tertiary structure. The validation of Protean establishes it as an unbiased and accurate method of generating a representative sampling of all the possible conformations which satisfy the experimental data. At the solid level, the precision is good enough to clearly define the topology of the protein. An analysis of Protean's performance using data sets of dismal to ideal quality permits us to define the limits of the precision with which a structure can be determined by this method. ------------------------------ Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Seminar - Fuzzy Relational Databases (SMU) Design of Similarity-Based (Fuzzy) Relational Databases Speaker: Bill P. Buckles, University of Texas, Arlington Location 315 SIC, Southern Methodist University Time: 2:00PM While the core of an expert system is its inference mechanism, a common component is a database or other form of knowledge representation. The authors have developed a variation of the relational database model in which data that is linguistic or inherently uncertain may be represented. The keystone concept of this representation is the replacement of the relationship " is equivalent to" with the relationship "is similar to". Similarity is defined in fuzzy set theory as an $n sup 2$ relationship over a domain D, |D| = n such that i. s(x,x)=1, x member D ii. s(x,y)=s(y,x) x,y member D iii. s(x,y) >= max[min(s(x,y),s(y,z))]; x, y, z member D Beginning with a universal relation, a method is given for developing the domain sets, similarity relationships and base relations for a similarity-based relational database. The universal relation itself enumerates all domains. The domain sets may be numeric (in which case no further design is needed) or scalar (in which case the selection of a comprehensive scalar set is needed). Similarity relationship contains $n sup 2$ values where n is the number of scalars in a domain set. A method is described for developing a set of consistent values when initially given n-1 values. The base relations are derived using fuzzy functional dependencies. This step also requires the identification of candidate keys. ------------------------------ Date: Fri 26 Sep 86 10:47:21-EDT From: Lisa F. Melcher Subject: Seminar - General Logic (MIT) Date: Thursday, October 2, 1986 Time: 1:45 p.m......Refreshments Time: 2:00 p.m......Lecture Place: NE43 - 512A " GENERAL LOGIC " Gordon Plotkin Department of Computer Science University of Edinburgh, Scotland A good many logics have been proposed for use in Computer Science. Implementing them involves repeating a great deal of work. We propose a general account of logics as regards both their syntax and inference rules. As immediate target we envision a system to which one inputs a logic obtaining a simple proof-checker. The ideas build on work in logic of Paulson, Martin-Lof and Schroeder-Heister and in the typed lambda-calculus of Huet and Coquand and Meyer and Reinhold. The slogan is: Judgements are Types. For example the judgement that a proposition is true is identified with its type of proofs; general and hypothetical judgements are identified with dependent product types. This gives one account of Natural Deduction. It would be interesting to extend the work to consider (two-sided) sequent calculi for classical and modal logics. Sponsored by TOC, Laboratory for Computer Science Albert Meyer, Host ------------------------------ Date: Fri 26 Sep 86 14:47:36-EDT From: Rosemary B. Hegg Subject: Seminar - Generic Tasks in Knowledge-Based Reasoning (MIT) Date: Wednesday, October 1, 1986 Time: 2.45 pm....Refreshments 3.00 pm....Lecture Place: NE43-512A GENERIC TASKS IN KNOWLEDGE-BASED REASONING: CHARACTERIZING AND DESIGNING EXPERT SYSTEMS AT THE ``RIGHT'' LEVEL OF ABSTRACTION B. CHANDRASEKARAN Laboratory for Artificial Intelligence Research Department of Computer and Information Science The Ohio State University Columbus, Ohio 43210 We outline the elements of a framework for expert system design that we have been developing in our research group over the last several years. This framework is based on the claim that complex knowledge-based reasoning tasks can often be decomposed into a number of @i(generic tasks each with associated types of knowledge and family of control regimes). At different stages in reasoning, the system will typically engage in one of the tasks, depending upon the knowledge available and the state of problem solving. The advantages of this point of view are manifold: (i) Since typically the generic tasks are at a much higher level of abstraction than those associated with first generation expert system languages, knowledge can be acquired and represented directly at the level appropriate to the information processing task. (ii) Since each of the generic tasks has an appropriate control regime, problem solving behavior may be more perspicuously encoded. (iii) Because of a richer generic vocabulary in terms of which knowledge and control are represented, explanation of problem solving behavior is also more perspicuous. We briefly describe six generic tasks that we have found very useful in our work on knowledge-based reasoning: classification, state abstraction, knowledge-directed retrieval, object synthesis by plan selection and refinement, hypothesis matching, and assembly of compound hypotheses for abduction. Host: Prof. Peter Szolovits ------------------------------ Date: Fri, 26 Sep 86 12:41:26 CDT From: forbus@p.cs.uiuc.edu (Kenneth Forbus) Subject: Conference - Workshop on Qualitative Physics Call for Participation Workshop on Qualitative Physics May 27-29, 1987 Urbana, Illinois Sponsored by: the American Association for Artificial Intelligence and Qualitative Reasoning Group University of Illinois at Urbana-Champaign Organizing Committee: Ken Forbus (University of Illinois) Johan de Kleer (Xerox PARC) Jeff Shrager (Xerox PARC) Dan Weld (MIT AI Lab) Objectives: Qualitative Physics, the subarea of artificial intelligence concerned with formalizing reasoning about the physical world, has become an important and rapidly expanding topic of research. The goal of this workshop is to provide an opportunity for researchers in the area to communicate results and exchange ideas. Relevant topics of discussion include: -- Foundational research in qualitative physics -- Implementation techniques -- Applications of qualitative physics -- Connections with other areas of AI (e.g., machine learning, robotics) Attendance: Attendence at the workshop will be limited in order to maximize interaction. Consequently, attendence will be by invitation only. If you are interested in attending, please submit an extended abstract (no more than six pages) describing the work you wish to present. The extended abstracts will be reviewed by the organizing committee. No proceedings will be published; however, a selected subset of attendees will be invited to contribute papers to a special issue of the International Journal of Artificial Intelligence in Engineering. There will be financial assistance for graduate students who are invited to attend. Requirements: The deadline for submitting extended abstracts is February 10th. On-line submissions are not allowed; hard copy only please. Any submission over 6 pages or rendered unreadable due to poor printer quality or microscopic font size will not be reviewed. Since no proceedings will be produced, abstracts describing papers submitted to AAAI-87 are acceptable. Invitations will be sent out on March 1st. Please send 6 copies of your extended abstracts to: Kenneth D. Forbus Qualitative Reasoning Group University of Illinois 1304 W. Springfield Avenue Urbana, Illinois, 61801 ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:43:13 1986 Date: Tue, 30 Sep 86 20:43:03 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #201 Status: R AIList Digest Monday, 29 Sep 1986 Volume 4 : Issue 201 Today's Topics: Bibliography - Definitions & Recent Articles in Robotics and Vision ---------------------------------------------------------------------- Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Defs for ai.bib35, new keyword code for editorials D MAG39 Computer Aided Design\ %V 18\ %N 3\ %D APR 1986 D MAG40 Automation and Remote Control\ %V 46\ %N 9 Part 2\ %D SEP 1985 D MAG41 IEEE Transactions on Industrial Electronics\ %V 33\ %N 2\ %D MAY 1986 D MAG42 Soviet Journal of Computer and Systems Sciences\ %V 23\ %N 6\ %D NOV-DEC 1985 D MAG43 Journal of Symbolic Computation\ %V 2\ %N 1\ %D MARCH 1986 D MAG44 Image and Vision Computing\ %V 3\ %N 4\ %D NOV 1985 D BOOK42 Second Conference on Software Development Tools, Techniques and Altern atives\ %I IEEE Computer Society Press\ %C Washington\ %D 1985 D BOOK43 Fundamentals of Computation Theory (Cottbus)\ %S Lecture Notes in Computer Science\ %V 199\ %I Springer-Verlag\ %C Berlin-Heidelberg-New York\ %D 1985 D BOOK44 Robot Sensors, Volume 1 (Vision)\ %I IFS Publications\ %C Bedford\ %D 1986 D BOOK45 Robot Sensors, Volume 2 (Tactile and Non-Vision)\ %I IFS Publications\ %C Bedford\ %D 1986 D MAG45 Journal of Logic Programming\ %V 2\ %D 1985\ %N 3 D BOOK46 Advances in Cryptology\ %S Lecture Notes in Computer Science\ %V 196\ %I Springer-Verlag\ %C Berlin-Heidelberg-New York\ %D 1985 D BOOK47 Mathematical Foundations of Software Development V 1\ %S Lecture Notes in Computer Science\ %V 185\ %I Springer-Verlag\ %C Berlin-Heidelberg-New York\ %D 1985 D MAG46 Proceedings of the 44th Session of the International Statistical Institute\ %V 1\ %D 1983 D BOOK48 Seminar Notes on Concurrency\ %S Lecture Notes in Computer Science\ %V 197\ %I Springer-Verlag\ %C Berlin-Heidelberg-New York\ %D 1985 D MAG47 Proceedings of the Conference "Algebra and Logic"\ %D 1984\ %C Zagreb D MAG48 Pattern Recognition\ %V 19\ %N 2\ %D 1986 D MAG49 IEEE Transactions on Geoscience and Remote Sensing\ %V 24\ %N 3\ %D MAY 1986 D MAG50 Information and Control\ %V 67\ %N 1-3\ %D OCT-DEC 1985 D MAG51 Kybernetes\ %V 15\ %N 2\ %D 1986 D MAG52 Data Processing\ %V 28\ %N 3\ %D APR 1986 D MAG53 J. Tsinghua Univ.\ %V 25\ %D 1985\ %N 2 D MAG54 Logique et. Anal (n. S.)\ %V 28\ %D 1985\ %N 110-111 D MAG55 Werkstattstechnik wt Zeitschrift fur Industrielle Fertigung\ %V 76\ %N 5\ %D MAY 1986 D MAG56 Robotica\ %V 4\ %D APR 1986 D MAG57 International Journal of Man Machine Studies\ %V 24\ %N 1\ %D JAN 1986 D MAG58 Computer Vision, Graphics and Image Processing\ %V 34\ %N 1\ %D APR 1986 D BOOK49 Flexible Manufacturing Systems: Methods and Studies\ %S Studies in Management Science and Systems\ %V 12\ %I North Holland Publishing Company\ %C Amsterdam\ %D 1986 D MAG59 International Journal for Robotics Research\ %V 5\ %N 1\ %D Spring 1986 D BOOK50 International Symposium on Logic Programming\ %D 1984 D MAG61 Proceedings of the 1986 Symposium on Symbolic and\ Algebriaic Computation\ %D JUL 21-23 1986 __________________________________________________________________________ A new keyword code for article types has been added, AT22, which is for editorials. ------------------------------ Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Recent Articles in Robotics and Vision %A Kunwoo Lee %A Daniel A. Tortorelli %T Computer-aided Design of Robotic Manipulators %J MAG39 %P 139-146 %K AI07 %A Ho Bin %T Inputting Constructive Solid Geometry Representations Directly from 2D Orthographic Engineering Drawings %J MAG39 %P 147-155 %K AA05 %A T. H. Richards %A G. C. Onwubolu %T Automatic Interpretation of Engineering Drawings for 3D Surface Representation in CAD %J MAG39 %P 156-160 %K AA05 %A J. S. Arora %A G. Baenziger %T Uses of Artificial Intelligence in Design Optimization %J Computer Methods in Mechanics and Engineering %V 54 %N 3 %D MAR 1986 %P 303-324 %K AA05 %A V. N. Burkov %A V. V. Tayganov %T Adaptive Functioning Mechanisms of Active Systems. I. Active Identification and Progressive Mechanisms %J MAG40 %P 1141-1146 %K AA20 AI09 AI04 AI08 AI13 %A A. A. Zlatopolskii %T Image Segmentation along Discontinuous Boundaries %J MAG40 %P 1160-1167 %K AI06 %A E. B. Yanovskaya %T Axiomatic Characterization of the Maxmin and the Lexicographic Maxmin Solution in Bargaining Schemes %J MAG40 %P 1177-1185 %K AI02 AI03 AA11 %A Yu. V. Malyshenko %T Estimating and Minimizing Diagnostic Information when Troubleshooting an Analog Device %J MAG40 %P 1192-1195 %K AA04 AA21 %A G. Hirzinger %T Robot Systems Completely Based on Sensory Feedback %J MAG41 %P 105-109 %K AI07 AI06 %A Y. Y. Hung %A S. K. Cheng %A N. K. Loh %T A Computer Vision Techniques for Surface Curvature Gaging with Project Grating %J MAG41 %P 158-161 %K AI07 AI06 %A Zvi Galil %T Optimal Prallel Algorithms for String Matching %J Information and Control %V 67 %N 1-3 %D 1985 %P 144-157 %K O06 %A E. Tanic %T Urban Planning and Artificial Intelligence - The Urbys System %J Computers, Environment and Urban Systems %V 10 %N 3-4 %D 1986 %P 135-146 %K AA11 %A B. M. Shtilman %T A Formal Linguistic Model for Solving Discrete Optimization Problems. II. The Language of Zones, Translations and the Boundary Problem %J MAG42 %P 17-28 %K AI02 AA05 %A V. A. Abramov %A A. I. Piskunov %A Yu. T. Rubanik %T A Modification to the Bellman-Zadeh Multistep Procedure for Decision Making under Fuzzy Conditions for Microelectronic Systems %J MAG42 %P 143-151 %K AI13 O05 %A James L. Eilbert %A Richard M. Salter %T Modeling Neural Networks in Scheme %J Simulation %V 46 %D 1986 %N 5 %P 193 %K AI12 T01 %A E. A. Shingareva %T Semiotic Basis of the Pragmatic Approach to Recognition of the Text Meaning %J Nauchno-Tekhnicheskaya Informatsiya, Seriya II- Informatisionnye Protessy I Sistemy %N 3 %D 1986 %K AI02 %A T. Kim %A K. Chwa %T Parallel Algorithms for a Depth First Search and a Breadth First Search %J International Journal of Computer Mathematics %V 19 %N 1 %D 1986 %P 39-56 %K AI03 H03 %A Hsu-Pin Wang %A Richard A. Wysk %T An Expert System for Machining Data Selection %J Computers and Industrial Engineering %V 10 %N 2 %D 1986 %K AA26 AI01 %A L. R. Rabiner %A F. K. Soong %T Single-Frame Vowel Recognition Using Vector Quantization with Several Distance Measures %J AT&T Technical Journal %V 64 %N 10 %D DEC 1985 %P 2319-2330 %K AI05 %A A. Pasztor %T Non-Standard Algorithmic and Dynamic Logics %J MAG43 %P 59-82 %A Alex P. Pentland %T On Describing Complex Surface Shapes %J MAG44 %P 153-162 %K AI06 AI16 %A B. F. Buxton %A D. W. Murray %T Optic Flow Segmentation as an Ill-posed and Maximum Likelihood Problem %J MAG44 %P 163-169 %K AI06 %A M. C. Ibison %A L. Zapalowski %A C. G. Harris %T Direct Surface Reconstruction from a Moving Sensor %J MAG44 %P 170-176 %K AI06 %A S. A. Lloyd %T Binary Stereo Algorithm Based on the Disparity-Gradient Limit and Using Optimization Theory %J MAG44 %P 177-182 %K AI06 %A Andrew Blake %A Andrew Zimmerman %A Greg Knowles %T Surface Descriptions from Stereo and Shading %J MAG44 %P 183-196 %K AI06 %A G. D. Sullivan %A K. D. Baker %A J. A D. W. Anderson %T Use of Multiple Difference-of-Gaussian Filters to Verify Geometric Models %J MAG44 %P 192-197 %K AI06 %A J. Hyde %A J. A. Fullwood %A D. R. Corrall %T An Approach to Knowledge Driven Segmentation %J MAG44 %P 198-205 %K AI06 %A J. Kittler %A J. Illingworth %T Relaxation Labelling Algorithm - A Review %J MAG44 %P 206-216 %K AI06 AT08 %A R. T. Ritchings %A A. C. F. Colchester %A H. Q. Wang %T Knowledge Based Analysis of Carotid Angiograms %J MAG44 %P 217 %K AI06 AA01 %A W. L. Mcknight %T Use of Grammar Templates for Software Engineering Environments %J BOOK42 %P 56-66 %K AA08 %A M. T. Harandi %A M. D. Lubars %T A Knowledge Based Design Aid for Software Systems %J BOOK42 %P 67-74 %K AA08 %A Y. Takefuji %T AI Based General Purpose Cross Assembler %J BOOK42 %P 75-85 %K AA08 %A R. N. Cronk %A D. V. Zelinski %T ES/AG System Generation Environment for Intelligent Application Software %J BOOK42 %P 96-100 %K AA08 %A B. Friman %T X - A Tool for Prototyping Through Examples %J BOOK42 %P 141-148 %K AA08 %A D. Hammerslag %A S. N. Kamin %A R. H. Campbell %T Tree-Oriented Interactive Processing with an Application to Theorem-Proving %J BOOK42 %P 199-206 %K AA08 AI11 %A Gudmund Frandsen %T Logic Programming and Substitutions %B BOOK43 %P 146-158 %K AI10 %A H. J. Cho %A C. K. Un %T On Reducing Computational Complexity in Connected Digit Recognition by the Frame Labeling Method %J Proceedings of the IEEE %V 74 %N 4 %D APR 1986 %P 614-615 %K AI06 %A Vijay Gehlot %A Y. N. Srikant %T An Interpreter for SLIPS - An Applicative Language Based on Lambda-Calculus %J Computer Languages %V 11 %N 1 %P 1-14 %D 1986 %A Sharon D. Stewart %T Expert System Invades Military %J Simulation %V 46 %N 2 %D FEB 1986 %P 69 %K AI01 AA18 %A F. C. Hadipriono %A H. S. toh %T Approximate Reasoning Models for Consequences on Structural Component Due to Failure Events %J Civil Engineering Pract Design Engineering %V 5 %N 3 %D 1986 %P 155-170 %K AA05 AA21 O04 %A J. Tymowski %T Industrial Robots %J Mechanik %V 58 %N 10 %D 1985 %P 493-496 %K AI07 %X (in Polish with English, Polish, Russian and German summaries) %A Dieter Schutt %T Expert Systems - Forerunners of a New Technology %J Siemens Review %V 55 %N 1 %D JAN- FEB 1986 %P 30 %K AI01 %A H. Kasamatu %A S. Omatu %T Edge-Preserving Restoration of Noisy Images %J International Journal of Systems Sciences %V 17 %N 6 %D JUN 1985 %P 833-842 %K AI06 %A A. Pugh %T Robot Sensors - A Personal View %B BOOK44 %P 3-14 %K AI07 %A L. J. Pinson %T Robot Vision - An Evaluation of Imaging Sensors %B BOOK44 %P 15-66 %K AI07 AI06 %A D. G. Whitehead %A I. Mitchell %A P. V. Mellor %T A Low-Resolution Vision Sensor %B BOOK44 %P 67-74 %K AI06 %A J. E. Orrock %A J. H. Garfunkel %A B. A. Owen %T An Integrated Vision/Range Sensor %B BOOK44 %P 75-84 %K AI06 %A S. Baird %A M. Lurie %T Precise Robotic Assembly Using Vision in the Hand %B BOOK44 %P 85-94 %K AI06 AI07 AA26 %A C. Loughlin %A J. Morris %T Line, Edge and Contour Following with Eye-in-Hand Vision %B BOOK44 %P 95-102 %K AI06 AI07 %A P. P. L. Regtien %A R. F. Wolffenbuttel %T A Novel Solid-State Colour Sensor Suitable for Robotic Applicatinos %B BOOK44 %P 103-114 %K AI06 AI07 %A A. Agrawal %A M. Epstein %T Robot Eye-in-Hand Using Fibre Optics %B BOOK44 %P 115-126 %K AI06 AI07 %A P. A. Fehrenbach %T Optical Alignment of Dual-in-Line Components for Assembly %B BOOK44 %P 127-138 %K AI06 AI07 AA26 AA04 %A Da Fa Li %T Semantically Positive Unit Resolution for Horn Sets %J MAG53 %P 88-91 %K AI10 %X Chinese with English Summary %A V. S. Neiman %T Proof Search without Repeated Examination of Subgoals %J Dokl. Akad. Nauk SSSR %V 286 %D 1986 %N 5 %P 1065-1068 %K AI11 %X Russian %A A. Colmerauer %T About Natural Logic. Automated Reasoning in Nonclassical Logic %J MAG54 %P 209-231 %K AI11 %A Ulf Grenander %T Pictures as Complex Systems %B Complexity, Language and Life: Mathematical Approaches %S Biomathematics %V 16 %I Spring %C Berlin-Heidelberg-New York %D 1986 %P 62-87 %K AI06 %A G. E. Mints %T Resolution Calculi for Nonclassical Logics %J Semiotics and Information Science %V 25 %P 120-135 %D 1985 %K AI11 %X Akad. Nauk SSSR, Vsesoyuz. Inst. Nauchn. i Tekn. Inform., Moscow (in Russian) %A Charles G. Margon %T Autologic. Automated Reasoning in Nonclassical Logic %J MAG54 %P 257-282 %K AI11 %A B. M. Shtilman %T A Formal Linguistic Model for Solving Discrete Optimization Problems I. Optimization tools. Language of Trajectories %J Soviet J. Computer Systems Science %V 23 %D 1985 %N 5 %P 53-64 %A David Lee %T Optimal Algorithms for Image Understanding: Current Status and Future Plans %J J. Complexity %V 1 %D 1985 %N 1 %P 138-146 %K AI06 %A Douglas B. West %A Prithviraj Banerjee %T Partial Matching in Degree-Restricted Bipartite Graphs %J Proceedings of the Sixteenth Southeastern International Conference on Combinatorics, Graph Theory and Computing %P 259-266 %D 1985 %K O06 %A Kyota Aoki %A N. Mugibayashi %T Cellular Automata and Coupled Chaos Developed in Lattice Chain of N Equivalent Switching Elements %J Phys. Lett. A %V 114 %D 1986 %N 8-9 %P 425-429 %K AI12 %A R. J. R. Back %T A Computational Interpretation of Truth Logic %J Synthese %V 66 %D 1986 %N 1 %P 15-34 %A Max Michel %T Computation of Tempral Operators: Automated Reasoning in Nonclassical Logic %J MAG54 %P 137-152 %K AI11 %A H. J. Warnecke %A A. Altenhein %T 2-1/2D Geometry Representation for Collision Avoidance of Industrial Robots %J MAG55 %P 269-272 %K AI07 %A W. Jacobi %T Industrial Robots - Already Sufficiently Flexible for the User %J MAG55 %P 273-277 %K AI07 %A H. J. Warnecke %A G. Schiele %T Measurement Methods for the Determination of Industrial Robot Characteristics %J MAG55 %P 278-280 %K AI07 %A H. H. Raab %T Assembly of Multipolar Plug Bonding Boxes in a Programmable Assembly Cell %J MAG55 %P 281-283 %K AA26 %A M. Schwiezer %A E. M. Wolf %T Strong Increase in Industrial Robot Installation %J MAG55 %P 286 %K AT04 AI07 %A T. W. Stacey %A A. E. Middleditch %T The Geometry of Machining for Computer-aided Manufacture %J MAG56 %P 83-92 %K AA26 %A S. S. Iyengar %A C. L. Jorgensen %A S. U. N. Rao %A C. R. Weisbin %T Robot Navigation Algorithms Using Learned Spatial Graphs %J MAG56 %P 93-100 %K AI07 %A Guy Jamarie %T On the Use of Time-Varying Intertia Links to Increase the Versatility of Manipulators %J MAG56 %P 101-106 %K AI07 %A Eugeny Krustev %A Ljubomir Lilov %T Kinematic Path Control of Robot Arms %J MAG56 %P 107-116 %K AI07 %A Tony Owen %T Robotics: The Strategic Issues %J MAG56 %P 117 %K AI07 %A C. H. Cho %T Expert Systems, Intelligent Devices, Plantwide Control and Self Tuning Algorithms: An Update on the ISA/86 Technical Program %J MAG56 %P 69 %K AA20 AI01 %A A. Hutchinson %T A Data Structure and Algorithm for a Self-Augmenting Heuristic Program %J The Computer Journal %P 135-150 %V 29 %N 2 %D APR 1986 %K AI04 %A B. Kosko %T Fuzzy Cognitive Maps %J MAG57 %P 65-76 %K AI08 O04 %A C. L. Borgman %T The Users Mental Model of an Information Retrieval System - An Experiment on a Prototype Online Catalog %J MAG57 %P 47-64 %K AI08 AA14 %A D. R. Peachey %A G. I. Mccalla %T Using Planning Techniques in Intelligent Tutoring Systems %J MAG57 %P 77 %K AA07 AI09 %A H. J. Bernstein %T Determining the Shape of a Convex n-sided Polygon Using 2n+k Tacticle Probes %J Information Processing Letters %V 22 %N 5 %D APR 28, 1986 %P 255-260 %K AI07 O06 %A Fu-Nian Ku %A Jian-Min Hu %T A New Approach to the Restoration of an Image Blurred by a Linear Uniform Motion %J MAG58 %P 20-34 %K AI06 %A Charles F. Neveu %A Charles R. Dyer %A Roland T. Chin %T Two-Dimensional Object Recognition Using Multiresolution Models %J MAG58 %P 52-65 %K AI06 %A Keith E. Price %T Hierarchical Matching Using Relaxation %J MAG58 %P 66-75 %K AI06 %A Angela Y. Wu %A S. K. Bhaskar %A Azriel Rosenfeld %T Computation of Geometric Properties from the Medial Axis Transform in O(n log n) Time %J MAG58 %P 76-92 %K AI06 O06 %A H. B. Bidasaria %T A Method for Almost Exact Historgram Matching for Two Digitized Images %J MAG58 %P 93-98 %K AI06 O06 %A Azriel Rosenfled %T "Expert" Vision Systems: Some Issues %J MAG58 %P 99-101 %K AI06 AI01 %A John R. Kender %T Vision Expert Systems Demand Challenging Expert Interactions %J MAG58 %P 102-103 %K AI06 AI01 %A Makoto Nagao %T Comment on the Position Paper \*QExpert Vision Systems\*U %J MAG58 %P 104 %K AI06 AI01 %A Leonard Uhr %T Workshop on Goal Directed \*QExpert\*U Vision Systems: My Positions and Comments %J MAG58 %P 105-108 %K AI06 AI01 %A William B. Thompson %T Comments on "Expert" Vision Systems: Some Issues %J MAG58 %P 109-110 %K AI06 AI01 %A V. A. Kovalevsky %T Dialog on "Expert" Vision Systems: Comments %J MAG58 %P 111-113 %K AI06 AI01 %A David Sher %T Expert Systems for Vision Based on Bayes Rule %J MAG58 %P 114-115 %K AI06 AI01 O04 %A S. Tanimoto %T The Case for Appropriate Architecture %J MAG58 %P 116 %K AI06 AI01 %A Azriel Rosenfeld %T Rosenfeld's Concluding Remarks %J MAG58 %P 117 %K AI06 AI01 %A Robert M. Haralick %T "Robot Vision" by Berthold Horn %J MAG58 %P 118 %K AI06 AI07 AT07 %A K. Shirai %A K. Mano %T A Clustering Experiment of the Spectra and the Spectral Changes of Speech to Extract Phonemic Features %J MAG58 %P 279-290 %K AI05 %A A. K. Chakravarty %A A. Shutub %T Integration of Assembly Robots in a Flexible Assembly System %B BOOK49 %P 71-88 %K AI07 AA26 %A R. C. Morey %T Optimizing Versatility in Robotic Assembly Line Design- An Application %B BOOK49 %P 89-98 %K AI07 AA26 %A J. Grobeiny %T The Simple Linguistic Approach to Optimization of a Plant Layout by Branch and Bound %B BOOK49 %P 141-150 %K AA26 AI02 AI03 %A Z. J. Czjikiewicz %T Justification of the Robots Applications %B BOOK49 %P 367-376 %K AI07 %A M. J. P. Shaw %A A. B. Whinston %T Applications of Artificial Intelligence to Planning and Scheduling in Flexible Manufacturing %B BOOK49 %P 223-242 %K AI07 %A S. Subramanymam %A R. G. Askin %T An Expert Systems Approach to Scheduling in Flexible Manufacturing Systems %B BOOK49 %P 243-256 %K AI07 %A Michael K. Brown %T The Extraction of Curved Surface Features with Generic Range Sensors %J MAG59 %P 3-18 %K AI06 %A Michael Erdmann %T Using Backprojections for Fine Motion Planning with Uncertaintly %J MAG59 %P 19-45 %K AI07 AI09 O04 %A Katsushi Ikeuchi %A H. Keith Nishihara %A Berthold K. P. Horn %A Patrick Sobalvarro %A Shigemi Nagati %T Determining Grasp Configurations Using Photometric Stereo and the PRISM Binocular Stereo System %J MAG59 %P 46-65 %K AI06 AI07 %A Dragan Stokic %A Miomir Vukobratovic %A Dragan Hristic %T Implementation of Force Feedback in Manipulation Robots %J MAG59 %P 66-76 %K AI07 %A Oussama Khatib %T Real-Time Obstacle Avoidance for Manipulators and Mobile Robots %J MAG59 %P 90-98 %K AI07 AA19 %A R. Featherstone %T A Geometric Investigation of Reach %J MAG59 %P 99 %K AI07 AT07 %A Maria Virginia Aponte %T Editing First Order Proofs: Programmed Rules vs. Derived Rules %J BOOK50 %P 92-98 %K AI11 %A Hellfried Bottger %T Automatic Theorem-Proving with Configuraitons %J Elektron. Informationsverarb. Kybernet. %V 21 %N 10-11 %P 523-546 %K AI11 %A D. R. Brough %A M. H. van Emden %T Dataflow, Flowcharts and \*QLUCID\*U style Programming in Logic %J BOOK50 %P 252-258 %A Laurent Fribough %T Handling Function Definitions Through Innermost Superposition and Rewriting %B BOOK30 %P 325-344 %A T. Gergely %A M. Szots %T Cuttable Formulas for Logic Programming %J BOOK50 %P 299-310 %A N. N. Leonteva %T Information Model of the Automatic Translation System %J Nauchno-Tekhnicheskaya Informatsiya, Seriya II - Informatsionnye Protsessy I Sistemy %N 10 %D 1985 %P 22-28 %X in Russian ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Sep 30 20:43:35 1986 Date: Tue, 30 Sep 86 20:43:21 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #202 Status: R AIList Digest Monday, 29 Sep 1986 Volume 4 : Issue 202 Today's Topics: Bibliography - Recent Reports ---------------------------------------------------------------------- Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Recent Reports %A Ru-qian Lu %T Expert Union: United Service of Distributed Expert Systems %R 85-3 %I University of Minnesota-Duluth %C Duluth, Minnesota %D June, 1985 %K H03 AI01 %X A scheme for connecting expert systems in a network called an {\nit expert union} is described. Consultation scheduling algorithms used to select the appropriate expert(s) to solve problems are proposed, as are strategies for resolving contradictions. %A J. C. F. M. Neves %A G. F. Luger %A L. F. Amaral %T Integrating a User's Knowledge into a Knowledge Base Using a Logic Based Representation %I University of New Mexico %R CS85-2 %K AA08 AI10 %A J. C. F. M. Neves %A G. F. Luger %T An Automated Reasoning System for Presupposition Analysis %I University of New Mexico %R CS85-3 %K AI16 %A J. C. F. M. Neves %A G. F. Luger %A J. M. Carvalho %T A Formalism for Views in a Logic Data Base %I University of New Mexico %R CS85-4 %K AA08 %A Franz Winkler %T A Note on Improving the Complexity of the Knuth-Bendix Completion Algorithm %I University of Delaware %R 85-04 %K AI14 %A Claudio Gutierrez %T An Integrated Office Environment Under the AI Paradigm %I University of Delaware %R 86-03 %K AA06 %A Amir M. Razi %T An Empirical Study of Robust Natural Language Processing %I University of Delaware %R 86-08 %K AI02 %A John T. Lund %T Multiple Cause Identification in Diagnostic Problem Solving %I University of Delaware %R 86-11 %K AA05 AA21 %A D. Nau %A T.C. Chang %T Hierarchical Representation of Problem-Solving Knowledge in a Frame-Based Process Planning System %I Production Automation Project, University of Rochester %R TM-50 %C Rochester, New York %K AA26 %T INEXACT REASONING IN PROLOG-BASED EXPERT SYSTEMS %A Koenraad G. Lecot %R CSD-860053 %I University of California, Los Angeles %K AI01 O04 T02 %$ 13.75 %X Expert systems are only worthy of their name if they can cope in a consistent and natural way with the uncertainty and vagueness that is inherent to real world expertise. This thesis explores the current methodologies, both in the light of their acceptabiity and of their implementation in the logic programming language Prolog. We treat in depth the subjective Bayesian approach to inexact reasoning and describe a meta-level implementation in Prolog. This probabilistic method is compared with an alternative theory of belief used in Mycin. We describe an implementation of Mycin's consultation phase. We argue further that the theory of fuzzy logic is more adequate to describe the uncertainty and vagueness of real world situations. Fuzzy logic is put in contrast with the probabilistic approaches and an implementation strategy is described. %T DISTRIBUTED DIAGNOSIS IN CAUSAL MODELS WITH CONTINUOUS VARIABLES %A Judea Pearl %R CSD-860051 %I University of California, Los Angeles %$ 1.50 %K O04 H03 AA21 %X We consider causal models in which the variables form a linearly coupled hierarchy, and are subject to Gaussian sources of noise. We show that if the number of circuits in the hierarchy is small, the impact of each new piece of evidence can be viewed as a perturbation that propagates through a network of processors (one per variable) by local communication. This mode of diagnosis admits flexible control strategies and facilitates the generation of intuitively meaningful explanations. %T RELAXATION PROBLEM SOLVING (with input to Chinese input problem) %A Kam Pui Chow %I University of California, Los Angeles %R CSD-860058 %$ 12.00 %K AI02 %X Two fundamental problem solving techniques are introduced to help automate the use of relaxation: multilevel frameworks and constraint generation. They are closely related to iterative relaxation and subproblem relaxation. .sp 1 In multilevel problem solving, the set of constraints is partitioned vertically into different levels. Lower level constraints generate possible solutions while higher level constraints prune the solutions to reduce the combinatorial explosion. Subproblem relaxation at first relaxes the high level constraints; the solution is then improved by strengthening the relaxed constraints. .sp 1 The constraint generation technique uses iterative relaxation to generate a set of constraints from a given model. This set of constraints with a constraint interpreter form an expert system. This is an improvement over most existing expert systems which require experts to write down their expertise in rules. .sp 1 These principles are illustrated by applying them to the Chinese input problem, which is to transform a phonetic spelling, without word breaks, of a Chinese sentence into the corresponding Chinese characters. Three fundamental issues are studied: segmentation, homophone analysis, and dictionary organization. The problem is partitioned into the following levels: phonetic spelling, word, and grammar. The corresponding constraints are legal spellings, legal words, and legal syntactic structures. Constraints for syntactic structure are generated from a Chinese grammar. %T RELAXATION PROCESSES: THEORY, CASE STUDIES AND APPLICATIONS %A Ching-Tsun Chou %R CSD-860057 %$ 6.25 %I University of California, Los Angeles %K O02 T02 AA08 %X Relaxation is a powerful problem-solving paradigm in coping with problems specified using constraints. In this Thesis we present a study of the nature of relaxation processes. We begin with identifying certain typical problems solvable by relaxation. Motivated by these concrete examples, we develop a formal theory of relaxation processes and design the General Relaxation Semi-Algorithm for solving general Relaxation Problems. To strengthen the theory, we do case studies on two relaxation-solvable problems: the Shortest-Path Problem and Prefix Inequalities. The principal results of these studies are polynomial-time algorithms for both problems. The practical usefulness of relaxation is demonstrated by implementing a program called TYPEINF which employs relaxation techniques to automatically infer types for Prolog programs. Finally we indicate some possible directions of future research. %A J. R. Endsor %A A. Dickinson %A R. L. Blumenthal %T Describe - An Explanation Facility for an Object Based System %I Carnegie Mellon Computer Science Department %D DEC 1985 %K AI01 O01 %A Kai-Fu Lee %T Incremental Network Generation in Template-Based Word Recognition %I Carnegie Mellon Computer Science Department %D DEC 1985 %K AI05 %A J. Quinlan %T A Comparative Analysis of Computer Architectures for Production System Machines %I Carnegie Mellon Computer Science Department %D MAY 1985 %K AI01 H03 OPS5 %A M. Boggs %A J. Carbonell %A M. Kee %A I. Monarch %T Dypar-I: Tutorial and Reference Manual %I Carnegie Mellon Computer Science Department %D DEC 1985 %K AI01 AI02 Franz Lisp %A Paola Giannini %T Type Checking and Type Deduction Techniques for Polymorphic Programming Languages %I Carnegie Mellon Computer Science Department %D DEC 1985 %K O02 lambda-calculus let construct %A M. Dyer %A M. Flowers %A S. Muchnick %T Lisp/85 User's Manual %I University of Kansas, Computer Science Department %R 77-4 %K T01 %A M. Flowers %A M. DYer %A S. Muchnick %T LISP/85 Implementation Report %I University of Kansas, Computer Science Department %R 78-1 %K T01 %A N. Jones %A S. Muchnick %T Flow Analysis and Optimization of LISP-like Structures %I University of Kansas, Computer Science Department %R 78-2 %K T01 %A U. Pleban %T The Standard Semantics of a Subset of SCHEME, A Dialect of LISP %I University of Kansas, Computer Science Department %R 79-3 %K T01 O02 %A S. Muchnick %A U. Pleban %T A Semantic Comparison of LISP and SCHEME %I University of Kansas, Computer Science Department %R 80-3 %K T01 O02 %A M. Jones %T The PEGO Acquisition System Implementaiton Report %I University of Kansas, Computer Science Department %R 80-4 %A Gary Borchardt %A Z. Bavel %T CLIP, Computer Language for Idea Processing %I University of Kansas, Computer Science Department %R 81-4 %A Marek Holynski %A Brian R. Gardner %A Rafail Ostrovsky %T Toward an Intelligent Computer Graphics System %I Boston University, Computer Science Department %R BUCS Tech Report #86-003 %D JAN 1986 %K T01 AA16 %A Joyce Friedman %A Carol Neidle %T Phonological Analysis for French Dictation: Preliminaries to an Intelligent Tutoring System %I Boston University, Computer Science Department %R BUCS Tech Report #86-004 %D APR 1986 %K AI02 AA07 %A Pawel Urzyczyn %T Logics of Programs with Boolean Memory %I Boston University, Computer Science Department %R BUCS Tech Report #86-006 %D APR 1986 %K AI16 %A Chua-Huang %A Christian Lengauer %T The Derivation of Systolic Implementatons of Programs %R TR-86-10 %I Department of Computer Sciences, University of Texas at Austin %D APR 1986 %K AA08 AA04 H03 H02 %A E. Allen Emerson %A Chin-Laung Lei %T Model Checking in the Propositional Mu-Calculus %R TR-86-06 %I Department of Computer Sciences, University of Texas at Austin %D FEB 1986 %K O02 AA08 %A R. D. Lins %T On the Efficiency of Categorical Combinators as a Rewriting System %D NOV 1985 %R No 34 %I University of Kent at Canterbury, Computing Laboratory %K AI11 AI14 %A R. D. Lints %T A Graph Reduction Machine for Execution of Categorical Combinators %D NOV 1985 %R No 36 %I University of Kent at Canterbury, Computing Laboratory %A S. J. Thompson %T Proving Properties of Functions Defined on Lawful Types %D MAY 1986 %R No 37 %I University of Kent at Canterbury, Computing Laboratory %K AA08 AI11 %A V. A. Saraswat %T Problems with Concurrent Prolog %D JAN 1986 %I Carnegie Mellon University, Department of Computer Science %K T02 H03 %A K. Shikano %T Text-Independent Speaker Recognition Expertiments using Codebooks in Vector quantization %D JAN 1986 %I Carnegie Mellon University %K AI05 %A S. Nakagawa %T Speaker Independent Phoneme Recognition in Continuous Speech by a Statistical Method and a Stochastic Dynamic Time Warping Method %D JAN 1986 %I Carnegie Mellon University %K AI05 %A F. Hau %T Two Designs of Functional Units for VLSI Based Chess Machines %D JAN 1986 %I Carnegie Mellon University %K AA17 H03 %X Brute force chess automata searching 8 piles (4 full moves) or deeper have been dominating the computer Chess scene in recent years and have reached master level performance. One intereting question is whether 3 or 4 additional piles couples with an improved evaluation scheme will bring forth world championship level performance. Assuming an optimistic branching ratio of 5, speedup of at least one hundred fold over the best current chess automaton would be necessary to reach the 11 or 12 piles per move range. %A Y. Iwasaki %A H. A. Simon %T Theories of Causual Ordering: Reply to de Kleer and Brown %D FEB 1986 %I Carnegie Mellon University %K Causality in Device Behavior AA04 %A H. Saito %A M. Tomita %T On Automatic Composition of Stereotypic Documents in Foreign Languages %D DEC 1985 %I Carnegie Mellon University %K AI02 %A T. Imielinski %T Query Processing in Deductive Databases with Incomplete Information %R DCS-TR-177 %I Rutgers University, Laboratory for Computer Science Research %K AA09 AI10 Horn Clauses Skolem functions %A T. Imielinski %T Abstraction in Query Processing %R DCS-TR-178 %I Rutgers University, Laboratory for Computer Science Research %K AA09 AI11 %A T. Imielinski %T Results on Translating Defaults to Circumscription %R DCS-TR-179 %I Rutgers University, Laboratory for Computer Science Research %K AA09 %A T. Imielinski %T Transforming Logical Rules by Relational Algebra %R DCS-TR-180 %I Rutgers University, Laboratory for Computer Science Research %K AA09 AI10 Horn clauses %A T. Imeielinski %T Automated Deduction in Databases with Incomplete Information %R DCS-TR-181 %I Rutgers University, Laboratory for Computer Science Research %K AA09 %A B. A. Nadel %T Representationi-Selection for Constraint Satisfaction Problems: A Case Study Using n-queens %R DCS-TR-182 %I Rutgers University, Laboratory for Computer Science Research %K AI03 AA17 %A B. A. Nadel %T Theory-Based Search Order Selection for Constraint Satisfaction Problems %R DCS-TR-183 %I Rutgers University, Laboratory for Computer Science Research %K AI03 %A C. V. Srinivasan %T Problems, Challenges and Opportunities in Naval Operational Planning %R DCS-TR-187 %I Rutgers University, Laboratory for Computer Science Research %K AI09 AA18 %A M. A. Bienkowski %T An Example of Structured Explanation Generation %I Princeton University Computer ScienceDepartment %D NOV 1985 %K O01 %A Bruce G. Buchanan %T Some Approaches to Knowledge Acquisition %I Stanford University Computer Science Department %R STAN-CS-85-1076 %D JUL 1985 %$ $5.00 %K AI16 %A John McCarthy %T Applications of Circumscription to Formalizing Common Sense Knowledge %I Stanford University Computer Science Department %R STAN-CS-85-1077 %D SEP 1985 %$ $5.00 %K AI15 %A Stuart Russell, Esq. %T The Compleat Guide to MRS %I Stanford University Computer Science Department %R STAN-CS-85-1080 %D JUN 1985 %$ $15.00 %K AI16 %A Jeffrey S. Rosenschein %T Rational Interaction: Cooperation among Intelligent Agents %I Stanford University Computer Science Department %R STAN-CS-85-1081 %D OCT 1985 %$ $15.00 %K AI16 %A Allen Van Gelder %T A Message Passing Framework for Logical Query Evaluation %I Stanford University Computer Science Department %R STAN-CS-85-1088 %D DEC 1985 %$ $5.00 %K AI10 Horn Clauses relational data bases H03 AA09 acyclic database schemas %A Jeffrey D. Ullman %A Allen Van Gelder %T Parallel Complexity of Logical Query Programs %I Stanford University Computer Science Department %R STAN-CS-85-1089 %D DEC 1985 %$ $5.00 %K AI10 H03 AA09 %A Kaizhi Yue %T Constructing and Analyzing Specifications of Real World Systems %I Stanford University Computer Science Department %R STAN-CS-86-1090 %D SEP 1985 %K AI01 AA08 %X available in microfilm only %A Li-Min Fu %T Learning Object-Level and Metal-Level Knowledge in Expert Systems %I Stanford University Computer Science Department %R STAN-CS-86-1091 %D NOV 1985 %$ $15.00 %K jaundice AI04 AI01 AA01 condenser %A Devika Subramanian %A Bruce G. Buchanan %T A General Reading List for Artificial Intelligence %I Stanford University Computer Science Department %R STAN-CS-86-1093 %D DEC 1985 %$ 10.00 %K AT21 %X bibliography for students studying for AI qualifying exam at Stanford %A Bruce G. Buchanan %T Expert Systems: Working Systems and the Research Literature %I Stanford University Computer Science Department %R STAN-CS-86-1094 %D DEC 1985 %$ 10.00 %K AT21 AI01 %A Jiawei Han %T Pattern-Based and Knowledge-Directed Query Compilation for Recursive Data Bases %I The University of Wisconsin-Madison Computer Sciences Department %R TR 629 %D JAN 1986 %$ 5.70 %K AA09 AI01 AI09 %X Abstract: Expert database systems (EDS's) comprise an interesting class of computer systems which represent a confluence of research in artificial intelligence, logic, and database management systems. They involve knowledge-directed processing of large volumes of shared information and constitute a new generation of knowledge management systems. Our research is on the deductive augmentation of relational database systems, especially on the efficient realization of recursion. We study the compilation and processing of recursive rules in relational database systems, investigating two related approaches: pattern-based recursive rule compilation and knowledge-directed recursive rule compilation and planning. Pattern-based recursive rule compilation is a method of compiling and processing recursive rules based on their recursion patterns. We classify recursive rules according to their processing complexity and develop three kinds of algorithms for compiling and processing different classes of recursive rules: transitive closure algorithms, SLSR wavefront algorithms, and stack-directed compilation algorithms. These algorithms, though distinct, are closely related. The more complex algorithms are generalizations of the simpler ones, and all apply the heuristics of performing selection first and utilizing previous processing results (wavefronts) in reducing query processing costs. The algorithms are formally described and verified, and important aspects of their behavior are analyzed and experimentally tested. To further improve search efficiency, a knowledge-directed recursive rule compilation and planning technique is introduced. We analyze the issues raised for the compilation of recursive rules and propose to deal with them by incorporating functional definitions, domain-specific knowledge, query constants, and a planning technique. A prototype knowledge-directed relational planner, RELPLAN, which maintains a high level user view and query interface, has been designed and implemented, and experiments with the prototype are reported and illustrated. %A A. P. Anantharman %A Sandip Dasgupta %A Tarak S. goradia %A Prasanna Kaikini %A Chun-Pui Ng %A Murali Subbarao %A G. A. Venkatesh %A Sudhanshu Verma %A Kumar A. Vora %T Experience with Crystal, Charlotte and Lynx %I The University of Wisconsin-Madison Computer Sciences Department %R TR 630 %D FEB 1986 %K H03 T02 Waltz constraint-propagation %X Abstract: This paper describes the most recent implementations of distributed algorithms at Wisconsin that use the Crystal multicomputer, the Charlotte operating system, and the Lynx language. This environment is an experimental testbed for design of such algorithms. Our report is meant to show the range of applications that we have found reasonable in such an environment and to give some of the flavor of the algorithms that have been developed. We do not claim that the algorithms are the best possible for these problems, although they have been designed with some care. In several cases they are completely new or represent significant modifications of existing algorithms. We present distributed implementations of B trees, systolic arrays, prolog tree search, the travelling salesman problem, incremental spanning trees, nearest-neighbor search in k-d trees, and the Waltz constraint-propagation algorithm. Our conclusion is that the environment, although only recently available, is already a valuable resource and will continue to grow in importance in developing new algorithms. %A William J, Rapaport %T SNePS Considered as a Fully Intensional Propositional Semantic Network %R TR 85-15 %I Univ. at Buffalo (SUNY), Dept. of Computer Science %D October 1985 %K Semantic Network Processing System, syntax, semantics, intensional knowledge representation system, cognitive modeling, database management, pattern recognition, expert systems, belief revision, computational linguistics aa01 ai09 ai16 %O 46 pages %X Price: $1.00 North America, $1.50 Other %A William J. Rapaport %T Logic and Artificial Intelligence %R TR 85-16 %I University at Buffalo (SUNY), Dept. of Computer Science %D November 1985 %K logic, propositional logic, predicate logic, belief systems AA16 %O 44 pages %X Price: $1.00 North America, $1.50 Other %A William J. Rapaport %T Review of "Ethical Issues in the Use of Computers" %R TR 85-17 %I University at Buffalo, Dept. of Computer Science %D November 1985 %K computer ethics O06 %O 6 pages %X Price: $1.00 North America, $1.50 Other %A Radmilo M. Bozinovic %T Recognition of Off-line Cursive Handwriting: a Case of Multi-level Machine Perception %I Univ. at Buffalo (SUNY), Dept. of Computer Science %D March 1985 %R TR 85-01 %K Cursive script recognition, artificial intelligence, computer vision, language perception, language understanding %O 150 pages %X Price: $2.00 North America, $3.00 other %A R. Hookway %T Verification of Abstract Types Whose Representation Share Storage %D April 1980 %I Case Western Reserve University, Computer Engineering and Science Department %R CES-80-02 %K AA09 %$ $2.00 %A G. Ernst %A J. K. Vavlakha %A W. F. Ogden %T Verification of Programs with Procedure-Type Parameters %I Case Western Reserve University, Computer Engineering and Science Department %R CES-80-11 %D 1980 %K AA09 %$ $2.00 %A G. Ernst %A F. T. Bradshaw %A R. J. Hookway %T A Note on Specifications of Concurrent Processes %I Case Western Reserve University, Computer Engineering and Science Department %R CES-81-01 %D FEB 1981 %K AA09 %$ $2.00 %A J. Franco %T The Probabilistic Analysis of the Pure Literal Heuristic in Theorem Proving %I Case Western Reserve University, Computer Engineering and Science Department %R CES-81-04 %D 1981 %K AI03 AI11 %$ $2.00 %A E. J. Branagan %T An Interactive Theorem Prover Verification %I Case Western Reserve University, Computer Engineering and Science Department %R CES-81-09 %D AUG 1981 %K AI11 %$ $2.00 %A G. W. Ernst %T A Method for verifying Concurrent Processes %I Case Western Reserve University, Computer Engineering and Science Department %R CES-82-01 %D FEB 1982 %K AA09 %$ $2.00 %A Chang-Sheng Yang %T A Computer Intelligent System for Understanding Chinese Homonyms %I Case Western Reserve University, Computer Engineering and Science Department %R CES-83-10 %D AUG 1983 %K AI02 %$ $2.00 %A G. Ernst %T Extensions to Methods for Learning Problem Solving Strategies %I Case Western Reserve University, Computer Engineering and Science Department %R CES-84-02 %D MAY 1984 %K AI04 %$ $2.00 %A R. J. Hookway %T Analysis of Asynchronous Circuits Using Temporal Logic %I Case Western Reserve University, Computer Engineering and Science Department %R CES-84-07 %D JUL 1984 %K AA04 %$ $2.00 %A Sterling, Leon %T Explaining Explanations Clearly %I Case Western Reserve University, Computer Engineering and Science Department %R CES-85-03 %D MAY 1985 %K O01 %$ $2.00 ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Oct 7 07:03:33 1986 Date: Tue, 7 Oct 86 07:03:27 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #203 Status: RO AIList Digest Monday, 6 Oct 1986 Volume 4 : Issue 203 Today's Topics: Query - Prolog Chemistry Systems & RuleMaster & AI Graduate Programs & Expert Systems and Deep Knowledge & Textbook for ES Applications & Communications Expert Systems & Generic Expert System & Integrated Inference Machines & Byte Prolog & Digitalk Smalltalk, AI Tools - Digitalk Smalltalk & Line Expert & XLISP &OPS5, Vision - Face Recognition, Logic Programming - TMS Loops ---------------------------------------------------------------------- Date: Sun, 28 Sep 86 10:46:15 -0200 From: Jacob Levy Subject: Chemistry systems & PROLOG Has anyone programmed or used a logic programming based system for use in Chemistry? I am especially interested in organic synthesis planning systems. Do you know of such systems written in other languages? Any help, references and info will be greatly appreciated, Thanks Rusty Red (AKA Jacob Levy) BITNET: jaakov@wisdom ARPA: jaakov%wisdom.bitnet@wiscvm.ARPA CSNET: jaakov%wisdom.bitnet@csnet-relay UUCP: jaakov@wisdom.uucp ------------------------------ Date: Sat, 27 Sep 86 09:15:20 cdt From: Esmail Bonakdarian Subject: RuleMaster Anybody out there have any comments about RuleMaster? RuleMaster (a product of the Radian Corporation) is a software tool for supporting the development of expert systems. I would be grateful for any information, comments from people who have used this package (especially on a DOS machine) etc. If there is enought interest I will collect and post all of the responses back to AIlist. Thanks, Esmail ------------------------------ Date: 29 Sep 86 00:29:16 GMT From: gatech!gitpyr!krubin@seismo.css.gov (Kenny Rubin) Subject: Differences among Grad AI programs The following is a request for information about the differences among the various universities that offer graduate degrees in AI. I apologize in advance if this topic has received prior discussion, I have been out the country for a few months and did not have access to the net. The goal of all this is to compile a current profile of the graduate AI programs at the different universities. Thus, any information about the different programs such as particular strengths and weaknesses would be useful. Also, a comparison and/or conclusions drawn between the various programs would be helpful. I am essentially interested in the areas of AI that each university performs research in. For example research pertaining to Knowledge Representation, Natural Language Processing, Expert System Development, Learning, Robotics, etc... Basically anything that you think potential applicants to the various universities would like to know, would be helpful. Feel free to comment about the university(ies) that you know best: - MIT, CMU, Yale, Standford, UC Berkeley, UCLA, etc... Please send all response by E-mail to me to reduce net traffic. If there is sufficient interest, I will post a compiled summary. Kenneth S. Rubin (404) 894-2348 Center for Man-Machine Systems Research School of Industrial and Systems Engineering Georgia Institute of Technology Post Office Box 35826 Atlanta, Georgia 30332 Majoring with: School of Information and Computer Science ...!{akgua,allegra,amd,hplabs,ihnp4,seismo,ut-ngp}!gatech!gitpyr!krubin ------------------------------ Date: 29 Sep 86 18:43:08 GMT From: mcvax!kvvax4!rolfs@seismo.css.gov (Rolf Skatteboe) Subject: Expert systems and deep knowledge Hello: For the time being, I'm working on my MSc thesis with main goal to investigate the combination of knowledge based diagnosis system and the use of mathematical models of gas turbines. I will used this models as deep knowledge in order to improve the results of the diagnosis system. The models can be used both as early warning fault systems and as sensor verification and test. The model can also be used to evaluate changes in machine parameters caused by engine degradiation. So far I have found some articles about diagnostic reasoning based on structure and behavior for digital electronic hardware. While I'm trying to find the best system structure for a demonstration system, I would like to get hold on information (articles references, program examples, and other people's experiences) both on using deep knowledge in expert systems in general, and the use of mathematical models in particular. I hope that someone can help me. Grethe Tangen Kongsberg KVATRO, NORWAY ------------------------------ Date: 3 Oct 1986 0904-EDT From: Holger Sommer Subject: Expert system Textbook For Applications I was asked to develop a course for Undergrad seniors and Beginning Graduated Students in Engineering, an introductory course for Expert System Technology with the focus on Application. I am looking for a suitable introductory textbook at the beginners level which could help me to get the students familiar with AI in general and expert systems specifically. Also if anyone has some experiance teachning a course for non-computer science students in the AI area I would appreciate our comments. Please send mail to Sommer@c.cs.cmu.edu ------------------------------ Date: Mon, 29 Sep 86 10:42:27 edt From: Lisa Meyer Subject: Request for Info on Expert Systems Development I am a senior Info & Computer Science major at Georgia Tech. I will be constructing an Expert System to diagnose communications setups & their problems for my senior design project at the request of my cooperative ed. employer. I have only had an introductory course in AI, so a large part of this project will be spent on researching information on expert system development. Any information on : Constructing Expert Systems (esp. for diagnostics) PC versions of Languages suitable for building Expert Systems Public Domain Expert Systems, ES Shells, or de- velopment tools Or good books, articles, or references to the subjects listed above WOULD BE GREATLY APPRECIATED. As the goal of my project is to con- struct a working diagnostic expert system and not to learn every- thing there is to know about AI, pointers to good sources of infromation, copies of applicable source, and information those who ARE knowledgable in the field of AI and Expert System Con- struction would be EXTREMELY HELPFUL. THANKS IN ADVANCE, Lisa Meyer (404-329-8022) Atlanta, GA ===================================================================== Lisa Meyer Harris / Lanier Computer R&D (Cooperative Education Program) Information & Computer Science Major Georgia Institute of Technology Georgia I Ga. Tech Box 30750, Atlanta Ga. 30332 {akgua,akgub,gatech}!galbp!lem ===================================================================== ------------------------------ Date: 30 Sep 86 13:34:16 GMT From: lrl%psuvm.bitnet@ucbvax.Berkeley.EDU Subject: Expert System Wanted Does anyone know of a general purpose expert system available for VM/CMS? I'm looking for one that would be used on a university campus by a variety of researchers in different disciplines. Each researcher would feed their own rules into it. Also, can anyone recommend readings, conferences, etc. for someone getting started in this field? Thanks. Linda Littleton phone: (814) 863-0422 214 Computer Building bitnet: LRL at PSUVM Pennsylvania State University University Park, PA 16802 ------------------------------ Date: Tue, 30 Sep 86 0:19:04 CDT From: Glenn Veach Subject: Address??? In a recent summary of the Spang-Robinson Report reference was made to the company "Integrated Inference Machines". Does anyone have an address for them? ------------------------------ Date: 2 Oct 86 18:21:29 GMT From: john@unix.macc.wisc.edu (John Jacobsen) Subject: pd prolog !!!!!!!!!!!!!! Does anyone have the public domain prolog package discussed in this month's BYTE magazine? John E. Jacobsen University of Wisconsin -- Madison Academic Computing Center ------------------------------ Date: 4 Oct 86 00:32:11 GMT From: humu!uhmanoa!todd@bass.nosc.mil (Todd Ogasawara) Subject: Digitalk Smalltalk for the PC If anyone out there has played with the version of Smalltalk for the PC by Digitalk, I'd like to get your opinions. I am especially interested in the object-oriented version of Prolog that comes with the package. Thanks..todd Todd Ogasawara, University of Hawaii Dept. of Psychology & U. of Hawaii Computing Center UUCP: {ihnp4,dual,vortex}!islenet! \ \__ uhmanoa!todd / {backbone}!sdcsvax!noscvax!humu!/ / clyde/ [soon to change to uhccux!todd] ARPA: humu!uhmanoa!todd@noscvax ** I used to be: ogasawar@nosc.ARPA & ogasawar@noscvax.UUCP ------------------------------ Date: 4 Oct 86 23:56:43 GMT From: spdcc!dyer@harvard.harvard.edu (Steve Dyer) Subject: Re: Digitalk Smalltalk for the PC I have it and am very impressed. Perhaps more convincing though, I have a friend who's been intimately involved with Smalltalk development from the very beginning who was also very impressed. It's even more remarkable because the Digitalk folks didn't license the Smalltalk-80 virtual machine from Xerox; they developed their system from the formal and not-so-formal specifications of Smalltalk 80 available in the public domain. Apparently, they can call their system "Smalltalk V" because "Smalltalk" isn't a trademark of Xerox; only "Smalltalk-80" is. I haven't played with their Prolog system written in Smalltalk. -- Steve Dyer dyer@harvard.HARVARD.EDU {linus,wanginst,bbnccv,harvard,ima,ihnp4}!spdcc!dyer ------------------------------ Date: 3 Oct 1986 13:30:37 EDT From: David Smith Subject: Communications Experts I was curious to find out about Werner Uhrig's question (9/10) relating to an Infoworld article from Smyrna, Ga since Ga is not exactly a hotbed of AI activity. I spoke to Nat Atwell of Concept Development Systems about Line Expert ($49.95). It is apparently an off-line Turbo Prolog application with knowledge about data set interfacing, known problems etc., including the ability to draw schematics of cables on the screen for you. For more info, call nat at (404) 434-4813. ------------------------------ Date: Fri, 26 Sep 86 11:06 PDT From: JREECE%sc.intel.com@CSNET-RELAY.ARPA Subject: XLISP Availability Although XLISP is available on a number of PC bulletin boards, your best bet for the PC version would be the BIX network run by Byte magazine. It has its own forum run by the author, David Betz, and you can turn around a message to him in 1-2 days. Information on how to sign up has been in most of the recent issues of Byte. Also, the latest version is 1.7, and there is talk of a compiler coming out in the future. John Reece Intel ------------------------------ Date: Mon, 29 Sep 86 0:30:53 BST From: Fitch@Cs.Ucl.AC.UK Subject: OPS5 on small machines (re V4 #183) There is OPS5 for the IBM-PC running UOLISP, from North West Computer Algorithms. It is the franz version slightly modified. I have run OPS5 on an Atari and on an Amiga. It does not need a very big system to do some things. ==John Fitch ------------------------------ Date: Mon 29 Sep 86 15:40:24-CDT From: Charles Petrie Reply-to: Petrie@MCC Subject: TMS Query Response More detail on Don Rose's TMS query: Does anyone know whether the standard algorithms for belief revision (e.g. dependency-directed backtracking in TMS-like systems) are guaranteed to halt? That is, is it possible for certain belief networks to be arranged such that no set of mutually consistent beliefs can be found (without outside influence)? There are at least three distinct Doyle-style algorithms. Doyle's doesn't terminate on unsatisfiable cicularities. James Goodwin's algorithm does. This algorithm is proved correct in "An Improved Algorithm for Non-monotonic Dependency Net Update", LITH-MAT-R-82-23, Linkoping Institute of Technology. David Russinoff's algorithm not only halts given an unsatisfiable circularity, but is guaranteed to find a well-founded, consistent set of status assignments, even if there are odd loops, if such a set is possible. There are dependency nets for which Russinoff's algorithm will properly assign statuses and Goodwin's may not. An example and proof of correctness for this algorithm is given in "An Algorithm for Truth Maintenance", AI-068-85, Microelectronics and Computer Technology Corporation. Also, Doyle made the claim that an unsatisfiable circularity can be detected if a node is its own ancestor after finding a valid justification with a NIL status in the Outlist. Detection of unsatisfiable circularities turns out to be more difficult than this. This is noted in "A Diffusing Computation for Truth Maintenance" wherein I give a distributed computation for status assignment (published in the Proc. 1986 Internat. Conf. on Parallel Processing, IEEE) that halts on unsatisfiable circularities. The term "unsatisfiable circularity" was introduced by Doyle and refers to a dependency network that has no correct status labeling. The term "odd loop" was introduced by Charniak, Riesbeck, and McDermott in section 16.7 of "Artificial Intelligence Programming". An equivalent definition is given by Goodwin. In both, an odd loop refers to a particular circular path in a dependency net. As Goodwin notes, such odd loops are a necessary, but not sufficient, condition for an unsatisfiable circularity. All of the algorithms mentioned above are for finding a proper set of status assignments for a dependency network. A distinct issue is the avoidance of the creation of odd loops, which may introduce unsatisfiable circularities, by Doyle-style dependency-directed backtracking. Examples of creation of such odd loops and algorithms to avoid such are described in my technical reports on DDB. Michael Reinfrank's report on the KAPRI system also notes the possibility of odd loops created by DDB. (DDB references on request to avoid an even longer note.) Charles Petrie PETRIE@MCC ------------------------------ Date: 3 Oct 1986 13:36:55 EDT From: David Smith Subject: Computer Vision Peter Snilovicz recently asked about recognizing faces. I saw a really interesting presentation on the subject Cortical Thought Theory by Rick Routh, ex-AFIT now with the Army at Fort Gordon. He can be reached at (404)791-3011. ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Oct 7 07:03:50 1986 Date: Tue, 7 Oct 86 07:03:39 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #204 Status: RO AIList Digest Monday, 6 Oct 1986 Volume 4 : Issue 204 Today's Topics: Seminars - Connectionist Networks (UPenn) & Automatic Class Formation (SRI) & Computers are not Omnipotent (CMU) & Automating Diagnosis (CMU) & Temporal Logic (MIT) & Program Transformations and Parallel Lisp (SU) & Temporal Theorem Proving (SU) & Efficient Unification of Quantified Terms (MIT) & Planning Simultaneous Actions (BBN) & Cognitive Architecture (UPenn) ---------------------------------------------------------------------- Date: Mon, 29 Sep 86 14:52 EDT From: Tim Finin Subject: Seminar - Connectionist Networks (UPenn) CONNECTIONIST NETWORKS Jerome A. Feldman Computer Science Department University of Rochester There is a growing interest in highly interconnected networks of very simple processing elements within artificial intelligence circles. These networks are referred to as Connectionist Networks and are playing an increasingly important role in artificial intelligence and cognitive science. This talk briefly discusses the motivation behind pursuing the the connectionist approach, and discusses a connectionist model of how mammals are able to deal with visual objects and environments. The problems addressed include perceptual constancies, eye movements and the stable visual world, object descriptions, perceptual generalizations, and the representation of extrapersonal space. The development is based on an action-oriented notion of perception. The observer is assumed to be continuously sampling the ambient light for information of current value. The central problem of vision is taken to be categorizing and locating objects in the environment. The critical step in this process is the linking of visual information to symbolic object descriptions, i.e., indexing. The treatment focuses on the different representations of information used in the visual system. The model employs four representation frames that capture information in the following forms: retinotopic, head-based, symbolic, and allocentric. The talk ends with a discussion of how connectionist models are being realized on existing architectures such as large multiprocessors. Thursday, October 2, 1986 Room 216 - Moore School 3:00 - 4:30 p.m. Refreshments Available Faculty Lounge - 2:00 - 3:00 p.m. ------------------------------ Date: Wed 1 Oct 86 11:46:40-PDT From: Amy Lansky Subject: Seminar - Automatic Class Formation (SRI) PROBABILISTIC PREDICTION THROUGH AUTOMATIC CLASS FORMATION Peter Cheeseman (CHEESEMAN@AMES-PLUTO) NASA Ames Research Center 11:00 AM, MONDAY, October 6 SRI International, Building E, Room EJ228 A probabilistic expert system is a set of probabilistic connections (e.g. conditional or joint probabilities) between the known variables. These connections can be used to make (conditional) probabilistic predictions for variables of interest given any combination of known variable values. Such systems suffer a major computational problem--- once the probabilstic connections form a complex inter-connected network, the cost of performing the necessary probability calculations becomes excessive. One approach to reducing the computational complexity is to introduce new "variables" (hidden causes or dummy nodes) that decouple the interactions between the variables. Judea Pearl has described an algorithm for introducing sufficient dummy nodes to create a tree structure, provided the probabilistic connections satisfy certain (strong) restrictions. This talk will describe a procedure for finding only the significant "hidden causes", that not only lead to a computationally simple procedure, but subsume all the significant interactions between the variables. VISITORS: Please arrive 5 minutes early so that you can be escorted up from the E-building receptionist's desk. Thanks! ------------------------------ Date: 28 Sep 1986 1228-EDT From: David A. Evans Subject: Seminar - Computers are not Omnipotent (CMU) PHILOSOPHY COLLOQUIUM ANNOUNCEMENT: COMPUTERS ARE NOT OMNIPOTENT David Harel Weizmann Institute and Carnegie Mellon University Monday, October 6 4:00 p.m. Porter Hall 223D In April, 1984, TIME magazine quoted a computer professional as saying: "Put the right kind of sofware into a computer and it will do whatever you want it to. There may be limits on what you can do with the machines themselves, but there are no limits on what you can do with the software." In the talk we shall disprove this contention outright, by exhibiting a wide array of results obtained by mathematicians and computer scientists between 1935 and 1983. Since the results point to inherent limitations of any kind of computing device, even with unlimited resources, they appear to have interesting philosophical implications concerning our own limitations as entities with finite mass. ------------------------------ Date: 29 September 1986 2247-EDT From: Masaru Tomita@A.CS.CMU.EDU Subject: Seminar - Automating Diagnosis (CMU) Date: 10/7 (Tuesday) Time: 3:30 Place: WeH 5409 Some AI Applications at Digital Automating Diagnosis: A case study Neil Pundit Kamesh Ramakrishna Artificial Intelligence Applications Group Digital Equipment Corporation 77 Reed Road (HLO2-3/M10) Hudson, Massachusetts, 01749 The Artificial Intelligence Applications Group at Digital is engaged in the development of expert systems technology in the context of many real-life problems drawn from within the corporation and those of customers. In addition, the group fosters basic research in AI by arrangements with leading universities. We plan to briefly describe some interesting applications. However, to satisfy your appetite for technical content, we will describe in some detail our progress on Beta, a tool for automating diagnosis. The communication structure level is a knowledge level at which certain kinds of diagnostic reasoning can occur. It is an intermediate level between the level at which current expert systems are designed (using knowledge acquired from experts) and the level at which ``deep reasoning'' systems perform (based on knowledge of structure, function, and behavior of the system being diagnosed). We present an example of an expert system that was designed the old-fashioned way and the heuristics that were the basis for recognizing the existence of the communication structure level. Beta is a language for specifying the communication structure of a system so that these heuristics can be compiled into a partially automatically generated program for diagnosing system problems. The current version of Beta can handle a specific class of communication structure that we call a ``control hierarchy'' and can analyze historical usage and error data maintained as a log file. The compiler embeds the heuristics in a generated mix of OPS5 and C code. We believe that Beta is a better way for designers and programmers who are not AI experts to express their knowledge of a system than the current rule-based or frame-based formalisms. ------------------------------ Date: Thu, 2 Oct 86 15:12:49 EDT From: "Elisha P. Sacks" Subject: Seminar - Temporal Logic (MIT) E. Taatnoon "The Third Cybernetics and Temporal Logic" I aim to link up the concepts of system bifurcation and system catastrophe with temporal logic in order to show the applicability of dialectical reasoning to metamorphic system transformations. A system catastrophe is an innovation resulting from reorganization resulting from a switch from positive to negative feedback or vice versa. The subsystems would then be oscillators and the truth of any descriptive statement is then distributive. Such oscillations would produce an uncertainty in the temporal trajectory of the system which would increase both towards the past and the future. This means that time is not a scalar dimension, but a quadratic paraboloid distribution of converging and diverging transition probabilities. A social system composed of such oscillators would be heterarchical rather than hierarchical. Refreshments. Hosts: Dennis Fogg and Boaz Ben-Zvi Place: 8th Floor Playroom Time: Noon ------------------------------ Date: 30 Sep 86 0947 PDT From: Carolyn Talcott Subject: Seminar - Program Transformations and Parallel Lisp (SU) Speaker: James M. Boyle, Argonne National Laboratory Time: Monday, October 6, 4pm Place: 252 Margaret Jacks (Stanford Computer Science Dept) Deriving Parallel Programs from Pure LISP Specifications by Program Transformation Dr. James M. Boyle Mathematics and Computer Science Division Argonne National Laboratory Argonne, IL 60439-4844 boyle@anl-mcs.arpa How can one implement a "dusty deck" pure Lisp program on global- memory parallel computers? Fortunately, pure Lisp programs have a declara- tive interpretation, which protects their decks from becoming too dusty! This declarative interpretation means that a pure Lisp program is not over-specified in the direction of sequential execution. Thus there is hope to detect parallelism automatically in pure Lisp programs. In this talk I shall describe a stepwise refinement of pure Lisp pro- grams that leads to a parallel implementation. From this point of view, the pure Lisp program is an abstract specification, which program transfor- mations can refine in several steps to a parallel program. I shall describe some of the transformations--correctness preserving rewrite rules --used to carry out the implementation. An important property of a parallel program is whether it can deadlock. I shall discuss a number of the design decisions involved in the refinement and their role in preserving the correctness of the transformed program, especially with regard to deadlock. Implementing a transformational refinement often leads to interesting insights about programming. I shall discuss some of these insights, including one about the compilation of recursive programs, and some that suggest ways to systematically relax the "purity" requirement on the Lisp program being implemented. We have used this approach to implement a moderately large pure Lisp program (1300 lines, 42 functions) on several parallel machines, including the Denelcor HEP (r.i.p.), the Encore Multimax, the Sequent Balance 8000, and the Alliant FX/8. I shall discuss some measurements of the performance of this program, which has achieved a speedup of 12.5 for 16 processors on realistic data, and some of the optimizations used to obtain this perfor- mance. Oh, yes, and by the way, the transformations produce a parallel pro- gram in FORTRAN! ------------------------------ Date: 01 Oct 86 1134 PDT From: Martin Abadi Subject: Seminar - Temporal Theorem Proving (SU) PhD Oral Examination Wednesday, October 8, 2:15 PM Margaret Jacks Hall 146 Temporal Theorem Proving Martin Abadi Computer Science Department In the last few years, temporal logic has been applied in the specification, verification, and synthesis of concurrent programs, as well as in the synthesis of robot plans and in the verification of hardware devices. Nevertheless, proof techniques for temporal logic have been quite limited up to now. This talk presents a novel proof system R for temporal logic. Proofs are generally short and natural. The system is based on nonclausal resolution, an attractive classical logic method, and involves a special treatment of quantifiers and modal operators. Unfortunately, no effective proof system for temporal logic is complete. We examine soundness and completeness issues for R and other systems. For example, a simple extension of our resolution system is as powerful as Peano Arithmetic. (Fortunately, refreshments will follow the talk.) Like classical resolution, temporal resolution suggests an approach to logic programming. We explore temporal logic as a programming language and a temporal resolution theorem prover as an interpreter for programs in this language. Other modal logics have found a variety of uses in artificial intelligence and in the analysis of distributed systems. We discuss resolution systems analogous to R for the modal logics K, T, K4, S4, S5, D, D4, and G. ------------------------------ Date: Sat, 4 Oct 86 12:30:41 EDT From: "Steven A. Swernofsky" Subject: Seminar - Efficient Unification of Quantified Terms (MIT) From: Susan Hardy JOHN STAPLES University of Queensland Efficient unification of quantified terms DATE: Tuesday, October 7, l986 TIME: 2:45 pm. - Refreshments 3:00 pm. - Talk PLACE: 2nd Floor Lounge Quantifiers such as for-all, integral signs, block headings would be a valuable enrichment of the vocabulary of a logic programming language or other computational logic. The basic technical prerequisite is a suitable unification algorithm. A programme is sketched for the development of data structures and algorithms which efficiently support the use of quantified terms. Progress in carrying out this programme is reviewed. Both structure sharing and non structure sharing representations of quantified terms are described, together with a unification algorithm for each case. The efficiency of the approach results from the techniques used to represent terms, which enable naive substitution to implement correct substitution for quantified terms. The work is joint with Peter J. Robinson. HOST: Arvind ------------------------------ Date: Sat, 4 Oct 86 13:04:33 EDT From: "Steven A. Swernofsky" Subject: Seminar - Planning Simultaneous Actions (BBN) From: Brad Goodman BBN Laboratories Science Development Program AI/Education Seminar Speaker: Professor James Allen University of Rochester (james@rochester) Title: Planning Simultaneous Actions in Temporally Rich Worlds Date: 10:30a.m., Monday, October 6th Location: 3rd floor large conference room, BBN Labs, 10 Moulton Street, Cambridge This talk describes work done with Richard Pelavin over the last few years. We have developed a formal logic of action that allows us to represent knowledge and reason about the interactions between events that occur simultaneously or overlap in time. This includes interactions between two (or more) actions that a single agent might perform simultaneously, as well as interactions between an agent's actions and events occuring in the external world. The logic is built upon an interval-based temporal logic extended with modal operators similar to temporal necessity and a counterfactual operator. Using this formalism, we can represent a wide range of possible ways in which actions may interact. ------------------------------ Date: Thu, 2 Oct 86 11:34 EDT From: Tim Finin Subject: Seminar - Cognitive Architecture (UPenn) WHAT IS THE SHAPE OF THE COGNITIVE ARCHITECTURE? Allen Newell Computer Science Department Carnegie Mellon University 12:00 noon, October 17 Alumni Hall, Towne Building University of Pennsylvania The architecture plays a critical role in computational systems, defining the separation between structure and content, and hence the capability of being programmed. All architectures have much in common. However, important characteristics depend on which mechanisms occur in the architecture (rather than in software) and what shape they take. There has been much research recently on architectures throughout computer and cognitive science. Within computer science the main drivers have been new hardware technologies (VLSI) and the felt need for parallelism. Within cognitive science the main drivers have been the hope of comprehensive psychological models (ACT*), the urge to ground the architecture in neurophysiological mechanisms (the connectionists) and the proposal of modularity as a general architectural principle (from linguistics). The talk will be on human cognitive architecture, but considerations will be brought to bear from everywhere. ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Oct 7 07:04:05 1986 Date: Tue, 7 Oct 86 07:03:51 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #205 Status: RO AIList Digest Monday, 6 Oct 1986 Volume 4 : Issue 205 Today's Topics: Humor - AI Limericks by Henry Kautz, AI Tools - Turbo Prolog & Reference Counts vs Garbage Collection, Philosophy - Emergent Consciousness & Perception ---------------------------------------------------------------------- Date: Fri, 3 Oct 86 14:58:37 CDT From: Glenn Veach Subject: AI Limericks by Henry Kautz gleaned from the pages of CANADIAN ARTIFICIAL INTELLIGENCE September 1986 No. 9 page 6: AI Limericks by Henry Kautz University of Rochester *** *** If you're dull as a napkin, don't sigh; Make your name as a "deep" sort of guy. You just have to crib, see Any old book by Kripke And publish in AAAI. *** *** A hacker who studied ontology Was famed for his sense of frivolity. When his program inferred That Clyde is a bird He blamed not his code but zoology. *** *** If your thesis is utterly vacuous Use first-order predicate calculus. With sufficient formality The sheerist banality Will be hailed by the critics: "Miraculous!" If your thesis is quite indefensible Reach for semantics intensional. Your committee will stammer Over Montague grammar Not admitting it's incomprehensible. ------------------------------ Date: Fri, 26 Sep 86 11:40 PDT From: JREECE%sc.intel.com@CSNET-RELAY.ARPA Subject: Turbo Prolog - Yet Another Opinion Although Turbo Prolog has been characterized by some wags as a "brain-dead implementation" I think its mixture of strengths and weaknesses would be more accurately described as those of an idiot savant. Some of the extensions, such as the built-in string editor predicates, are positively serendipitous, and you get most of the development time advantages of a fifth generation language for a conventional application plus good runtime performance for only $70. On the other hand, one tires quickly of writing NP-incomplete sets of type declarations which are unnecessary in any other implementation.... If nothing else, for $70 you can prototype something that can be used to justify spending $700 for a real PC Prolog compiler, or $18,000 for a VAX implementation. John Reece Intel ------------------------------ Date: Fri, 26 Sep 86 18:52:26 CDT From: neves@ai.wisc.edu (David M. Neves) Reply-to: neves@ai.wisc.edu (David M. Neves) Subject: Re: Xerox vs Symbolics -- Reference counts vs Garbage collection When I was using MIT Lisp Machines (soon to become Symbolics) years ago nobody used the garbage collector because it slowed down the machine and was somewhat buggy. Instead people operated for hours/days until they ran out of space and then rebooted the machine. The only time I turned on the garbage collector was to compute 10000 factorial. Do current Symbolics users use the garbage collector? "However, it is apparent that reference counters will never reclaim circular list structure." This is a common complaint about reference counters. However I don't believe there is very many circular data structures in real Lisp code. Has anyone looked into this? Has any Xerox user run out of space because of circular data structures in their environment? -- David Neves, Computer Sciences Department, University of Wisconsin-Madison Usenet: {allegra,heurikon,ihnp4,seismo}!uwvax!neves Arpanet: neves@rsch.wisc.edu ------------------------------ Date: 26 Sep 86 15:35:00 GMT From: princeton!siemens!steve@CAIP.RUTGERS.EDU Subject: Garb Collect Symb vs Xerox I received mail that apparently also went onto the net, from Dan Hoey (hoey@nrl-aic.ARPA). He discussed garbage collection in response to my unsupported allegation that, "S[ymbolics] talks about their garbage collection more, but X[erox]'s is better." I am glad to see someone taking up an informed discussion in this area. First, I briefly recap his letter, eliding (well-put) flames: + In the language of computer + science, Xerox reclaims storage using a ``reference counter'' + technique, rather than a ``garbage collector.'' + If we are to believe Xerox, the reference counter + technique is fundamentally faster, and reclaims acceptable amounts of + storage. However, it is apparent that reference counters will never + reclaim circular list structure. As a frequent user of circular list + structure (doubly-linked lists, anyone?), I find the lack tantamount to + a failure to reclaim storage. + I have never understood why Xerox continues to neglect to write a + garbage collector. It is not necessary to stop using reference counts, + but simply to have a garbage collector available for those putatively + rare occasions when they run out of memory. + Dan Hoey Xerox's system is designed for highly interactive use on a personal workstation (sound familiar?). They spread the work of storage reclamation evenly throughout the computation by keeping reference counts. Note that they have many extra tricks such as "References from the stack are not counted, but are handled separately at "sweep" time; thus the vast majority of data manipulations do not cause updates to [the reference counts]" (Interlisp-D Reference Manual, October, 1985). Even if this scheme were to use a greater total amount of CPU time than typical garbage collection, it would remain more acceptable for use on a personal, highly interactive workstation. I have no idea how it can be compared to Symbolics for overall performance, without comparing the entire Interlisp vs. Zetalisp systems. Nevertheless, I can say that my experience is that Interlisp runs a "G.C." every few seconds and it lasts, subjectively, an eyeblink. Occasionally I get it to take longer, for example when I zero my pointers to 1500 arrays in one fell swoop. I have some figures from one application, too. An old, shoddy implementation ran 113 seconds CPU and 37.5 seconds GC (25% GC). A decent implementation of the same program, running a similar problem twice, got 145 seconds CPU, but 10.8 and 20.3 seconds GC (6.9% and 12% GC). (The good implementation still doesn't have a good hashing function so it's still slower.) I cannot claim that these figures are representative. I have heard horror stories about other Lisps' GCs, although I don't have any feel for Symbolics's "Ephemeral GC". I have a strong feeling Xerox has other tricks besides the one about the stack, which they don't want to tell anyone. I know they recently fixed the reference counter from counting 16 or more references as "infinity" (and thus never reclaimable) to an overflow scheme where the reference count gets squirreled away somewhere else when it gets bigger. Finally, normally the amount of unreclaimed garbage (e.g. circular lists) grows much slower than memory fragments, so you have to rebuild your world before unreclaimed garbage becomes a problem anyway. Postfinally, Xerox makes a big deal that their scheme takes time proportional to the number of objects reclaimed, while traditional systems take time proportional to the number of objects allocated. I think Symbolics's ephemeral scheme is a clever way to consider only subsets of the universe of allocated objects, that are most likely to have garbage. I wish I knew whether it is a band-aid or an advance in the state-of-the-art. Absolutely ultimately, "traditional" GC I refer to, is known as "mark- and-sweep". Steve Clark {topaz or ihnp4}!princeton!siemens!steve ------------------------------ Date: Mon, 29 Sep 86 14:07:12 edt From: segall@caip.rutgers.edu (Ed Segall) Subject: Re: Emergent Consciousness Why must we presume that the seat of consciousness must be in the form of neural "circuits"? What's to prevent it from being a symbolic, logical entity, rather than a physical entity? After all, the "center of control" of most computers is some sort of kernal program, running on the exact same hardware as the other programs. (Don't try to push the analogy too far, you can probably find a hole in it.) Perhaps the hierarchical system referred to is also not structural. Might the brain operate even more like a conventional computer than we realize, taking the role of an extremely sophisticated (self-modifying) interpreter? The "program" that is interpreted is the pattern of firings occurring at any given time. If this is so, then moment-to-moment thought is almost completely in terms of the dynamic information contained in neural signals, rather than the quasi-static information contained in neural interconnections. The neurons simply serve to "run" the thoughts. This seems obvious to me, since I am assuming that neural firings can process information much faster than structural changes in neurons. I'd be interested to know about what rate neuron firings occur in the brain, and if anyone has an intelligent guess as to how much information can be stored at once in the "dynamic" form of firings rather than the "static" form of interconnections. I apologize in advance if what I suggest goes against well-understood knowlege (not theory) of how the brain operates. My information is from the perspective of a lay person, not a cognitive scientist. ------------------------------ Date: Mon, 29 Sep 86 09:34:01 EDT From: "Col. G. L. Sicherman" Subject: Re: Consciousness as bureaucracy Ken Laws' analogy between Bureaucracy and Man--more precisely, Man's Mind--has been anticipated by Marvin Minsky. I do not have the reference; I think it was a rather broad article in a general science journal. As I recall, the theory that Minsky proposed lay somewhere between the lay concept of self and the Zen concept. It seemed to suggest that consciousness is an illusion to itself, but a genuine and observable phenomenon to an outside observer, characterizable with the metaphor of bureaucracy. Perhaps some Ailist reader can identify the article. Emergent consciousness has always been a hope of A.I. I side with those who suggest that consciousness depends on contact with the world ... even though I know some professors who seem to be counter-examples! :-) ------------------------------ Date: 2 Oct 86 17:14:00 EDT From: "FENG, THEO-DRIC" Reply-to: "FENG, THEO-DRIC" Subject: Perception I just ran across the following report and thought it might contribute some- thing to the discussion on the "perception" of reality. (I'll try to summarize the report where I can.) according to Thomas Saum in the German Research Service, Special Press Reports, Vol. II, No. 7/86 A group of biologists in Bremen University has been furthering the theory devel- oped by Maturana and Varela (both from Chile) in the late '70's, that the brain neither reflects nor reproduces reality. They suggest that the brain creates its own reality. Gerhard Roth, a prof. of behavioral physiology at Bremen (with doctorates in philosophy and biology), has written several essays on the subject. In one, he ...writes that in the past the "aesthesio-psychological perspective" of the psychomatic problem was commonly held by both laypersons and scientists. This train of thought claims that the sensory organs reporduce the world at least partially and convey this image to the brain, where it is then reassembled ("reconstructed") in a uniform concept. In other words, this theory maintains that the sense organs are the brain's gateway to the world. In order to illustrate clearly the incorrectness of this view, Roth suggests that the perspectives be exchanged: if one looks at the problem of perception from the brain's angle, instead of the sense organs, the brain merely receives uniform and basically homo- geneous bioelectric signals from the nerve tracks. It is capable of determining the intensity of the sensory agitation by the frequency of these signals, but this is all it can do. The signals provide no information on the quality of the stimulation, for instance, on whe- ther an object is red or green. Indeed, they do not even say any- thing about the modality of the stimulus, i.e. whether it is an optical, acoustical, or chemical stimulation. The constructivists [as these new theoreticians are labeled], believe that the brain is a self-contained system. Its only access to the world consists of the uniform code of the nerve signals which have nothing in common with the original stimuli. Since the brain has no original image, it cannot possibly "reporduce" reality; it has to create it itself. "It (the brain) has to reconstruct the di- versity of the outside owrld from the uniform language of the neu- rons", Roth claims. The brain accomplishses this task by "interpret- ing itself", i.e. by construing what is going on inside itself. Thus, the brain "draws conclusions" from the degree to which it is agitated by the modality of the original stimulus: all neuronal im- pulses reaching the occipital cortex, for example, are visual im- pressions. This isolated nature of the brain and its reality, however, are by no means a blunder on the part of nature; indeed, they are not even a necessary evil, Roth explains. On the contrary, it is an adaptive advantage acquired by more highlly developed creatures dur- ing the course of their phylogenic development. If the brain had di- rect access to the environment, Roth argues, then one and the same stimulus would necessarily always result in one and the same reac- tion by the organizsm. Since, however, the human brain has retained a certain amount of creative scope for its reconstruction of reality, it is in a position to master complicated stiuations and adapt itself to unforeseen circumstances. Only in this way is it possible to recognize an object in differ- ent light intensities, from a new angle of vision, or at a distance. Even experiments with "reversal spectacles" demonstrate man's powers of adaptation in interpreting reality: after a little while, test persons, who see the world upside down with special glasses, simply turn their environment around again in their "mind". When, after a few days, they remove the spectacles, the "real" world suddenly seems to be standing on its head. This mobility and adaptability on the part of our perceptive fa- culties were obviously much more important for the evolution of more highly developed vertebrates than was a further intensification of the signal input by the sense organs. The million fibers in man's optic nerve are only double the number of a frog's; the human brain, on the other hand, has one hundred thousand times more nerve cells than a frog brain. But first and foremost, the "reality workshop", i.e., the cerebral area not tied to specific sense, has expanded during the evolution of man's brain, apparently to the benefit of our species. Contact: Prof. Dr. Gerhard Roth, Forschungsschwerpunkt Biosystemforschung, Universitat [note: umlaut the 'a'] Bremen, Postfach 330 440, D-2800 Bremen 33, West Germany. [conveyed by Theo@ARI] ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Oct 7 07:04:25 1986 Date: Tue, 7 Oct 86 07:04:11 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #206 Status: RO AIList Digest Monday, 6 Oct 1986 Volume 4 : Issue 206 Today's Topics: Philosophy - Searle, Turing, Symbols, Categories ---------------------------------------------------------------------- Date: 27 Sep 86 14:20:21 GMT From: princeton!mind!harnad@caip.rutgers.edu (Stevan Harnad) Subject: Searle, Turing, Symbols, Categories The following are the Summary and Abstract, respectively, of two papers I've been giving for the past year on the colloquium circuit. The first is a joint critique of Searle's argument AND of the symbolic approach to mind-modelling, and the second is an alternative proposal and a synthesis of the symbolic and nonsymbolic approach to the induction and representation of categories. I'm about to publish both papers, but on the off chance that there is a still a conceivable objection that I have not yet rebutted, I am inviting critical responses. The full preprints are available from me on request (and I'm still giving the talks, in case anyone's interested). *********************************************************** Paper #1: (Preprint available from author) MINDS, MACHINES AND SEARLE Stevan Harnad Behavioral & Brain Sciences 20 Nassau Street Princeton, NJ 08542 Summary and Conclusions: Searle's provocative "Chinese Room Argument" attempted to show that the goals of "Strong AI" are unrealizable. Proponents of Strong AI are supposed to believe that (i) the mind is a computer program, (ii) the brain is irrelevant, and (iii) the Turing Test is decisive. Searle's point is that since the programmed symbol-manipulating instructions of a computer capable of passing the Turing Test for understanding Chinese could always be performed instead by a person who could not understand Chinese, the computer can hardly be said to understand Chinese. Such "simulated" understanding, Searle argues, is not the same as real understanding, which can only be accomplished by something that "duplicates" the "causal powers" of the brain. In the present paper the following points have been made: 1. Simulation versus Implementation: Searle fails to distinguish between the simulation of a mechanism, which is only the formal testing of a theory, and the implementation of a mechanism, which does duplicate causal powers. Searle's "simulation" only simulates simulation rather than implementation. It can no more be expected to understand than a simulated airplane can be expected to fly. Nevertheless, a successful simulation must capture formally all the relevant functional properties of a successful implementation. 2. Theory-Testing versus Turing-Testing: Searle's argument conflates theory-testing and Turing- Testing. Computer simulations formally encode and test models for human perceptuomotor and cognitive performance capacities; they are the medium in which the empirical and theoretical work is done. The Turing Test is an informal and open-ended test of whether or not people can discriminate the performance of the implemented simulation from that of a real human being. In a sense, we are Turing-Testing one another all the time, in our everyday solutions to the "other minds" problem. 3. The Convergence Argument: Searle fails to take underdetermination into account. All scientific theories are underdetermined by their data; i.e., the data are compatible with more than one theory. But as the data domain grows, the degrees of freedom for alternative (equiparametric) theories shrink. This "convergence" constraint applies to AI's "toy" linguistic and robotic models as well, as they approach the capacity to pass the Total (asympototic) Turing Test. Toy models are not modules. 4. Brain Modeling versus Mind Modeling: Searle also fails to note that the brain itself can be understood only through theoretical modeling, and that the boundary between brain performance and body performance becomes arbitrary as one converges on an asymptotic model of total human performance capacity. 5. The Modularity Assumption: Searle implicitly adopts a strong, untested "modularity" assumption to the effect that certain functional parts of human cognitive performance capacity (such as language) can be be successfully modeled independently of the rest (such as perceptuomotor or "robotic" capacity). This assumption may be false for models approaching the power and generality needed to pass the Total Turing Test. 6. The Teletype versus the Robot Turing Test: Foundational issues in cognitive science depend critically on the truth or falsity of such modularity assumptions. For example, the "teletype" (linguistic) version of the Turing Test could in principle (though not necessarily in practice) be implemented by formal symbol-manipulation alone (symbols in, symbols out), whereas the robot version necessarily calls for full causal powers of interaction with the outside world (seeing, doing AND linguistic understanding). 7. The Transducer/Effector Argument: Prior "robot" replies to Searle have not been principled ones. They have added on robotic requirements as an arbitrary extra constraint. A principled "transducer/effector" counterargument, however, can be based on the logical fact that transduction is necessarily nonsymbolic, drawing on analog and analog-to-digital functions that can only be simulated, but not implemented, symbolically. 8. Robotics and Causality: Searle's argument hence fails logically for the robot version of the Turing Test, for in simulating it he would either have to USE its transducers and effectors (in which case he would not be simulating all of its functions) or he would have to BE its transducers and effectors, in which case he would indeed be duplicating their causal powers (of seeing and doing). 9. Symbolic Functionalism versus Robotic Functionalism: If symbol-manipulation ("symbolic functionalism") cannot in principle accomplish the functions of the transducer and effector surfaces, then there is no reason why every function in between has to be symbolic either. Nonsymbolic function may be essential to implementing minds and may be a crucial constituent of the functional substrate of mental states ("robotic functionalism"): In order to work as hypothesized, the functionalist's "brain-in-a-vat" may have to be more than just an isolated symbolic "understanding" module -- perhaps even hybrid analog/symbolic all the way through, as the real brain is. 10. "Strong" versus "Weak" AI: Finally, it is not at all clear that Searle's "Strong AI"/"Weak AI" distinction captures all the possibilities, or is even representative of the views of most cognitive scientists. Hence, most of Searle's argument turns out to rest on unanswered questions about the modularity of language and the scope of the symbolic approach to modeling cognition. If the modularity assumption turns out to be false, then a top-down symbol-manipulative approach to explaining the mind may be completely misguided because its symbols (and their interpretations) remain ungrounded -- not for Searle's reasons (since Searle's argument shares the cognitive modularity assumption with "Strong AI"), but because of the transdsucer/effector argument (and its ramifications for the kind of hybrid, bottom-up processing that may then turn out to be optimal, or even essential, in between transducers and effectors). What is undeniable is that a successful theory of cognition will have to be computable (simulable), if not exclusively computational (symbol-manipulative). Perhaps this is what Searle means (or ought to mean) by "Weak AI." ************************************************************* Paper #2: (To appear in: "Categorical Perception" S. Harnad, ed., Cambridge University Press 1987 Preprint available from author) CATEGORY INDUCTION AND REPRESENTATION Stevan Harnad Behavioral & Brain Sciences 20 Nassau Street Princeton NJ 08542 Categorization is a very basic cognitive activity. It is involved in any task that calls for differential responding, from operant discrimination to pattern recognition to naming and describing objects and states-of-affairs. Explanations of categorization range from nativist theories denying that any nontrivial categories are acquired by learning to inductivist theories claiming that most categories are learned. "Categorical perception" (CP) is the name given to a suggestive perceptual phenomenon that may serve as a useful model for categorization in general: For certain perceptual categories, within-category differences look much smaller than between-category differences even when they are of the same size physically. For example, in color perception, differences between reds and differences between yellows look much smaller than equal-sized differences that cross the red/yellow boundary; the same is true of the phoneme categories /ba/ and /da/. Indeed, the effect of the category boundary is not merely quantitative, but qualitative. There have been two theories to explain CP effects. The "Whorf Hypothesis" explains color boundary effects by proposing that language somehow determines our view of reality. The "motor theory of speech perception" explains phoneme boundary effects by attributing them to the patterns of articulation required for pronunciation. Both theories seem to raise more questions than they answer, for example: (i) How general and pervasive are CP effects? Do they occur in other modalities besides speech-sounds and color? (ii) Are CP effects inborn or can they be generated by learning (and if so, how)? (iii) How are categories internally represented? How does this representation generate successful categorization and the CP boundary effect? Some of the answers to these questions will have to come from ongoing research, but the existing data do suggest a provisional model for category formation and category representation. According to this model, CP provides our basic or elementary categories. In acquiring a category we learn to label or identify positive and negative instances from a sample of confusable alternatives. Two kinds of internal representation are built up in this learning by "acquaintance": (1) an iconic representation that subserves our similarity judgments and (2) an analog/digital feature- filter that picks out the invariant information allowing us to categorize the instances correctly. This second, categorical representation is associated with the category name. Category names then serve as the atomic symbols for a third representational system, the (3) symbolic representations that underlie language and that make it possible for us to learn by "description." This model provides no particular or general solution to the problem of inductive learning, only a conceptual framework; but it does have some substantive implications, for example, (a) the "cognitive identity of (current) indiscriminables": Categories and their representations can only be provisional and approximate, relative to the alternatives encountered to date, rather than "exact." There is also (b) no such thing as an absolute "feature," only those features that are invariant within a particular context of confusable alternatives. Contrary to prevailing "prototype" views, however, (c) such provisionally invariant features MUST underlie successful categorization, and must be "sufficient" (at least in the "satisficing" sense) to subserve reliable performance with all-or-none, bounded categories, as in CP. Finally, the model brings out some basic limitations of the "symbol-manipulative" approach to modeling cognition, showing how (d) symbol meanings must be functionally anchored in nonsymbolic, "shape-preserving" representations -- iconic and categorical ones. Otherwise, all symbol interpretations are ungrounded and indeterminate. This amounts to a principled call for a psychophysical (rather than a neural) "bottom-up" approach to cognition. ------------------------------ Date: Mon 29 Sep 86 09:55:11-PDT From: Pat Hayes Subject: Searle's logic I try not to get involved in these arguments, but bruce krulwich's assertion that Searle 'bases all his logic on' the binary nature of computers is seriously wrong. We could have harware which worked with direct, physical, embodiments of all of Shakespeare, and Searles arguments would apply to it just as well. What bothers him ( and many other philosophers ) is the idea that the machine works by manipulating SYMBOLIC descriptions of its environment ( or whatever it happens to be thinking about ). It's the internal representation idea, which we AIers take in with our mothers milk, which he finds so silly and directs his arguments against. Look, I also don't think there's any real difference between a human's knowledge of a horse and machine's manipulation of the symbol it is using to represent it. But Searle has some very penetrating arguments against this idea, and one doesnt make progress by just repeating one's intuitions, one has to understand his arguments and explain what is wrong with them. Start with the Chinese room, and read all his replies to the simple counterarguments as well, THEN come back and help us. Pat Hayes ------------------------------ Date: 1 Oct 86 18:25:16 GMT From: cbatt!cwruecmp!cwrudg!rush@ucbvax.Berkeley.EDU (rush) Subject: Re: Searle, Turing, Symbols, Categories (Question not comment) In article <158@mind.UUCP> harnad@mind.UUCP (Stevan Harnad) writes: >6. The Teletype versus the Robot Turing Test: > >For example, the "teletype" (linguistic) version of the Turing... > whereas the robot version necessarily >calls for full causal powers of interaction with the outside >world (seeing, doing AND linguistic understanding). > Uh...I never heard of the "robot version" of the Turing Test, could someone please fill me in?? I think that understanding the reasons for such a test would help me (I make no claims for anyone else) make some sense out of the rest of this article. In light of my lack of knowledge, please forgive my presumption in the following comment. >7. The Transducer/Effector Argument: > >A principled >"transducer/effector" counterargument, however, can be based >on the logical fact that transduction is necessarily >nonsymbolic, drawing on analog and analog-to-digital >functions that can only be simulated, but not implemented, >symbolically. > [ I know I claimed no commentary, but it seems that this argument depends heavily on the meaning of the term "symbol". This could be a problem that only arises when one attempts to implement some of the stranger possibilities for symbolic entities. ] Richard Rush - Just another Jesus freak in computer science decvax!cwruecmp!cwrudg!rush ------------------------------ Date: 2 Oct 86 16:05:28 GMT From: princeton!mind!harnad@caip.rutgers.edu (Stevan Harnad) Subject: Re: Searle, Turing, Symbols, Categories (Question not comment) In his commentary-not-reply to my <158@mind.UUCP>, Richard Rush <150@cwrudge.UUCP> asks: (1) > I never heard of the "robot version" of the Turing Test, > could someone please fill me in? He also asks (in connection with my "transducer/effector" argument) about the analog/symbolic distinction: (2) > I know I claimed no commentary, but it seems that this argument > depends heavily on the meaning of the term "symbol". This could > be a problem that only arises when one attempts to implement some > of the stranger possibilities for symbolic entities. In reply to (1): The linguistic version of the turing test (turing's original version) is restricted to linguistic interactions: Language-in/Language-out. The robotic version requires the candidate system to operate on objects in the world. In both cases the (turing) criterion is whether the system can PERFORM indistinguishably from a human being. (The original version was proposed largely so that your judgment would not be prejudiced by the system's nonhuman appearance.) On my argument the distinction between the two versions is critical, because the linguistic version can (in principle) be accomplished by nothing but symbols-in/symbols-out (and symbols in between) whereas the robotic version necessarily calls for non-symbolic processes (transducer, effector, analog and A/D). This may represent a substantive functional limitation on the symbol-manipulative approach to the modeling of mind (what Searle calls "Strong AI"). In reply to (2): I don't know what "some of the stranger possibilities for symbolic entities" are. I take symbol-manipulation to be syntactic: Symbols are arbitrary tokens manipulated in accordance with certain formal rules on the basis of their form rather than their meaning. That's symbolic computation, whether it's done by computer or by paper-and-pencil. The interpretations of the symbols (and indeed of the manipulations and their outcomes) are ours, and are not part of the computation. Informal and figurative meanings of "symbol" have little to do with this technical concept. Symbols as arbitrary syntactic tokens in a formal system can be contrasted with other kinds of objects. The ones I singled out in my papers were "icons" or analogs of physical objects, as they occur in the proximal physical input/output in transduction, as they occur in the A-side of A/D and D/A transformations, and as they may function in any part of a hybrid system to the extent that their functional role is not merely formal and syntactic (i.e., to the extent that their form is not arbitrary and dependent on convention and interpretation to link it to the objects they "stand for," but rather, the link is one of physical resemblance and causality). The category-representation paper proposes an architecture for such a hybrid system. Stevan Harnad princeton!mind!harnad ------------------------------ End of AIList Digest ******************** From csnet_gateway Tue Oct 7 18:47:03 1986 Date: Tue, 7 Oct 86 18:46:55 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #207 Status: R AIList Digest Tuesday, 7 Oct 1986 Volume 4 : Issue 207 Today's Topics: Seminars - Cross-Talk in Mental Operations (UCB) & Deductive Databases (UPenn) & Concept Acquisition in Noisy Environments (SRI) & Prolog without Horns (CMU) & Knowledge Engineering and Ontological Structure (SU), Conferences - AAAI-87 Tutorials & 1st Conf. on Neural Networks & Workshop on Qualitative Physics ---------------------------------------------------------------------- Date: Mon, 6 Oct 86 15:38:02 PDT From: admin%cogsci.Berkeley.EDU@berkeley.edu (Cognitive Science Program) Subject: Seminar - Cross-Talk in Mental Operations (UCB) BERKELEY COGNITIVE SCIENCE PROGRAM Cognitive Science Seminar - IDS 237A Tuesday, October 14, 11:00 - 12:30 2515 Tolman Hall Discussion: 12:30 - 1:30 2515 Tolman Hall ``Cross-Talk and Backward Processing in Mental Operations'' Daniel Kahneman Psychology Department There are many indications that we only have imperfect control of the operations of our mind. It is common to compute far more than is necessary for the task at hand. An operation of cleaning-up and inhibition of inappropriate responses is often required, and this operation is often only partially suc- cessful. For example, we cannot stop ourselves from reading words that we attend to; when asked to assess the similarity of two objects in a specified attribute we apparently compute many similarity relations in addition to the requisite one. The prevalence of such cross-talk has significant implications for a psychologically realistic notion of meaning for the interpre- tation of incoherence in judgments. A standard view of cognitive function is that the objects and events of expeience are assimilated, more or less success- fully, to existing schemas and expectations. Some perceptual and cognitive phenomena seem to fit another model, in which objects and events elicit their own context and define their own alternatives. Surprise, for example, is better viewed as a failure to make sense of an event post hoc than as a violation of expectations. Some rules by which events evoke counterfac- tual alternatives to themselves will be described. ------------------------------ Date: Sun, 5 Oct 86 11:15 EDT From: Tim Finin Subject: Seminar - Deductive Databases (UPenn) 3:00pm, Tuesday, October 7, 1986 23 Moore School, University of Pennsylvania EFFICIENT DEDUCTIVE DATABASES WILL THEY EVER BE CONSTRUCTED? Tomasz Imielinski Rutgers University The area of deductive databases is a rapidly growing field concerned with enhancing traditional relational databases with automated deduction capabilities. Because of the large amounts of data involved here the complexity issues become critical. We present a number of results related to the complexity of query processing in the deductive databases, both with complete and incomplete information. In an attempt to answer the question of whether efficient deductive databases will ever be constructed we demonstrate an idea of the "deductive database of the future". In such a system the concept of an answer to a query is tailored to the various limitations of computational resources. ------------------------------ Date: Mon 6 Oct 86 16:25:34-PDT From: Joani Ichiki Subject: Seminar - Concept Acquisition in Noisy Environments (SRI) L. Saitta (Dipartimento di Informatica, Universita di Torino, Italy) will present his talk entitled, "AUTOMATED CONCEPT ACQUISITION IN NOISY ENVIRONMENTS," 10/7/86 in EK242 at 11:00am. Abstract follows. This paper presents a system which performs automated concept acquisition from examples and has been especially designed to work in errorful and noisy environments. The adopted learning methodology is aimed to the target problem of finding discriminant descriptions of a given set of concepts and both examples and counterexamples are used. The learning knowledge is expressed in the form of production rules, organized into separate clusters, linked together in a graph structure; the condition part of the rules, corresponding to descriptions of relevant aspects of the concepts, is expressed by means of a first order logic based language, enriched with constructs suitable to handle uncertainty and vagueness and to increase readability by a human user. A continuous-valued semantics is associated to this language and each rule is affected by a certainty factor. Learning is considered as a cyclic process of knowledge extraction, validation and refinement; the control of the cycle is left to the teacher. Knowledge extraction proceeds through a process of specialization, rather than generalization, and utilizes a technique of problem reduction to contain the computational complexity. Moreover, the search strategy is strongly focalized by means of task-oriented but domain-independent heuristics, trying to emulate the learning mechanism of a human being, faced to find discrimination rules from a set of examples. Several criteria are proposed for evaluating the acquired knowledge; these criteria are used to guide the process of knowledge refinement. The methodology has been tested on a problem in the field of speech recognition and the obtained experimental results are reported and discussed. ------------------------------ Date: 6 October 1986 1411-EDT From: Peter Andrews@A.CS.CMU.EDU Subject: Seminar - Prolog without Horns (CMU) The following talk will be given in the Seminar on Automated Reasoning Wednesday, Oct. 15, at 4:30p.m. in room PH125C. The talk is independent of preceding material in the seminar. Prolog without Horns D. W. Loveland An extension to Prolog is defined that handles non-Horn clause sets (programs) in a manner closer to standard Prolog than previously proposed. Neither the negation symbol or a symbol for false are formally introduced in the system, although the system is conjectured to be propositionally complete. The intention of the extension is to provide processing of "nearly Horn" programs with minimal deviation from the Prolog format. Although knowledge of Prolog is not essential, some prior exposure to Prolog will be helpful. ------------------------------ Date: Mon 6 Oct 86 16:55:52-PDT From: Lynne Hollander Subject: Seminar - Knowledge Engineering and Ontological Structure (SU) SIGLUNCH Title: KNOWLEDGE ENGINEERING AS THE INVESTIGATION OF ONTOLOGICAL STRUCTURE Speaker: Michael J. Freiling Computer Research Laboratory Tektronix Laboratories Place: Chemistry Gazebo Time: 12:05-1:15, Friday, October 10 Experience has shown that much of the difficulty of learning to build knowledge-based systems lies in designing representation structures that adequately capture the necessary forms of knowledge. Ontological analysis is a method we have found quite useful at Tektronix for analyzing and designing knowledge-based systems. The basic approach of ontological analysis is a step-by-step construction of knowledge structures beginning with simple objects and relationships in the task domain, and continuing through representations of state, state transformations, and heuristics for selecting transformations. Formal tools that can be usefully employed in ontological analysis include domain equations, semantic grammars, and full-scale specification languages. The principles and tools of ontological analysis are illustrated with actual examples from knowledge-based systems we have built or analyzed with this method. ------------------------------ Date: Mon 29 Sep 86 10:39:41-PDT From: William J. Clancey Subject: AAAI-87 Tutorials AAAI-87 Tutorials -- Request for Proposals Tutorials will be presented at AAAI-87/Seattle on Monday, Tuesday, and Thursday, July 13, 14, and 16. Anyone interested in presenting a tutorial on a new or standard topic should contact the Tutorial Chair, Bill Clancey. Topic suggestions from tutorial attendees are also welcome. Potential speakers should submit a brief resume covering relevant background (primarily teaching experience) and any available examples of work (ideally, a published tutorial-level article on the subject). In addition, those people suggesting a new or revised topic should offer a 1-page summary of the idea, outlining the proposed subject and depth of coverage, identifying the necessary background, and indicating why it is felt that the topic would be well attended. With regard to new courses, please keep in mind that tutorials are intended to provide dissemination of reasonably well-agreed-upon information, that is, there should be a substantial body of accepted material. We especially encourage submission of proposals for new advanced topics, which in 1986 included "Qualitiative Simulation," "AI Machines," and "Uncertainty Management." Decisions about topics and speakers will be made by November 1. Speakers should be prepared to submit completed course material by December 15. Bill Clancey Stanford Knowledge Systems Laboratory 701 Welch Road, Building C Palo Alto, CA 94304 Clancey@SUMEX ------------------------------ Date: Tue, 30 Sep 86 11:43:56 pdt From: mikeb@nprdc.arpa (Mike Blackburn) Subject: 1st Conf. on Neural Networks CONFERENCE ANNOUNCEMENT: FIRST ANNUAL INTERNATIONAL CONFERENCE ON NEURAL NETWORKS San Diego, California 21-24 June 1987 The San Diego IEEE Section welcomes neural network enthusiasts in industry, academia, and government world-wide to participate in the inaugural annual ICNN conference in San Diego. Papers are solicited on the following topics: * Network Architectures * Learning Algorithms * Self- Organization * Adaptive Resonance * Dynamical Network Stability * Neurobiological Connections * Cognitive Science Connections * Electrical Neurocomputers * Opti- cal Neurocomputers * Knowledge Processing * Vision * Speech Recognition & Synthesis * Robotics * Novel Applications Contributed Papers: Extended Abstract should be submitted by 1 February 1987 for Conference Presentation. The Abstract must be single spaced, three to four pages on 8.5 x 11 inch paper with 1.5 inch margins. Abstracts will be carefully refereed. Accepted abstracts will be distributed at the conference. Final Papers due 1 June 1986. FINAL RELEASE OF ABSTRACTS AND PAPERS WITH RESPECT TO PROPRIETARY RIGHTS AND CLASSIFICATION MUST BE OBTAINED BEFORE SUBMITTAL. Address all Corresspondence to: Maureen Caudill - ICNN 10615G Tierrasanta Blvd. Suite 346, San Diego, CA 92124. Registration Fee: $350 if received by 1 December 1986, $450 thereafter. Conference Venue: Sheraton Harbor Island Hotel (approx. $95 - single), space limited, phone (619) 291-6400. Other lodg- ing within 10 minutes. Tutorials and Exhibits: Several Tutorials are Planned. Ven- dor Exhibit Space Available - make reservations early. Conference Chairman: Stephen Grossberg International Chairman: Teuvo Kohonen Organizing Committee: Kunihiko Fukushima, Clark Guest, Robert Hecht-Nielsen, Morris Hirsch, Bart Kosko (Chairman 619-457-5550), Bernard Widrow. September 30, 1986 ------------------------------ Date: 5 Oct 1986 13:16 EDT (Sun) From: "Daniel S. Weld" Subject: Workshop on Qualitative Physics Call for Participation Workshop on Qualitative Physics May 27-29, 1987 Urbana, Illinois Sponsored by: the American Association for Artificial Intelligence and Qualitative Reasoning Group University of Illinois at Urbana-Champaign Organizing Committee: Ken Forbus (University of Illinois) Johan de Kleer (Xerox PARC) Jeff Shrager (Xerox PARC) Dan Weld (MIT AI Lab) Objectives: Qualitative Physics, the subarea of artificial intelligence concerned with formalizing reasoning about the physical world, has become an important and rapidly expanding topic of research. The goal of this workshop is to provide an opportunity for researchers in the area to communicate results and exchange ideas. Relevant topics of discussion include: -- Foundational research in qualitative physics -- Implementation techniques -- Applications of qualitative physics -- Connections with other areas of AI (e.g., machine learning, robotics) Attendance: Attendence at the workshop will be limited in order to maximize interaction. Consequently, attendence will be by invitation only. If you are interested in attending, please submit an extended abstract (no more than six pages) describing the work you wish to present. The extended abstracts will be reviewed by the organizing committee. No proceedings will be published; however, a selected subset of attendees will be invited to contribute papers to a special issue of the International Journal of Artificial Intelligence in Engineering. Requirements: The deadline for submitting extended abstracts is February 10th. On-line submissions are not allowed; hard copy only please. Since no proceedings will be produced, abstracts describing papers submitted to AAAI-87 are acceptable. Invitations will be sent out on March 1st. Please send 6 copies of your extended abstracts to: Kenneth D. Forbus Qualitative Reasoning Group University of Illinois 1304 W. Springfield Avenue Urbana, Illinois, 61801 ------------------------------ End of AIList Digest ******************** From csnet_gateway Fri Oct 10 02:46:30 1986 Date: Fri, 10 Oct 86 02:46:21 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #208 Status: R AIList Digest Thursday, 9 Oct 1986 Volume 4 : Issue 208 Today's Topics: Bibliography - News and Recent Articles ---------------------------------------------------------------------- Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: News and Recent Articles %A Paul A. Eisenstein %T Detroit Finds Robots Aren't Living Up to Expectations %J Investor's Daily %D April 21, 1986 %P 12 %K AI07 Chrysler General Motors AA25 %X Chrysler said that automation was one of the major reasons productivity doubled since 1980. GM's Lake Orion, a "factory of the future" with 157 automated robots instead of providing the best quality and productivity of any GM plant is providing the lowest. Two other plants have been giving GM the same problems. %A Mary Petrosky %T Expert Software Aids Large System Design %J InfoWorld %D FEB 17, 1986 %V 8 %N 7 %P 1+ %K AA08 AI01 H01 AT02 AT03 Arthur Young Knowledge Ware %X Knowledge-Ware is selling the Information Engineering Workbench which provides tools to support developing business programs. It has features for supporting entity diagrams, data flow diagrams, etc. I cannot find any indication from this article where AI is actually used. %A John Gantz %T No Market Developing for Artificial Intelligence %J InfoWorld %D FEB 17, 1986 %V 8 %N 7 %P 27 %K AT04 AT14 %X D. M. Data predicts that the market for AI software will be $605 million this year and $2.65 billion in 1990. Arthur D. Little says it might be twice this. He argues that when you look at the companies, most of them are selling primarily to research market and not to the commercial data processing market. Intellicorp had 3.3 million in revenues for the 1984- 1985 fiscal year and it made a profit. However, a full third of its systems go to academics and 20 percent goes to Sperry for use in its own AI labs. %A Jay Eisenlohr %T Bug Debate %J InfoWorld %D FEB 17, 1986 %V 8 %N 7 %P 58 %K AT13 AT12 Airus AI Typist AT03 %X Response to harsh review of AI Typist by Infoworld from an employee of the company selling it. %A Eddy Goldberg %T AI offerings Aim to Accelerate Adoption of Expert Systems %J Computerworld %D MAY 26, 1986 %V 20 %N 21 %P 24 %K Teknowledge Carnegie Group Intel Hypercube Gold Hill Common Lisp AT02 H03 T03 T01 %X Teknowledge has rewritten S.1 in the C language. Intel has introduced Concurrent Common Lisp for its hypercube based machine %T New Products/Microcomputers %J Computerworld %D MAY 26, 1986 %V 20 %N 21 %P 94 %K AT04 AI06 H01 Digital Vision Computereyes %X Digital Vision introduced Computereyes video acquisition system for IBM PC. Cost is $249.95 without camera and $529.95 with one. %T New Products/Software and Services %J Computerworld %D MAY 26, 1986 %V 20 %N 21 %P 90 %K T03 AT02 %X LS/Werner has introduced a package containg four expert system tools for $1995. A guide to AI is also included. %A Douglas Barney %T AT&T Conversant Systems Unveils Voice Recognition Model %J ComputerWorld %D APR 21, 1986 %V 20 %N 16 %P 13 %K AI05 AT02 %X AT&T Conversant systems has two products to do speech recognition, the Model 80 which handles 80 simultaneous callers for $50,000 to $100,000 while the Model 32 costs between $25,000 and $50,000 and handles 32 simultaneous callers. It handles "yes," "no" and the numbers zero through nine. %A Charles Babcock %A James Martin %T MSA Users Give High Marks, Few Dollars to Information Expert %J ComputerWorld %D APR 21, 1986 %V 20 %N 16 %P 15 %K AA06 AT03 %X MSA has a product called Information Expert which integrates a variety of business applications through a shared dictionary and also provides reporting. However the 'expert system components' failed to live up to the "standard definition of expert systems." %A Alan Alper %T IBM Trumpets Experimental Speech Recognition System %J ComputerWorld %D APR 21, 1986 %V 20 %N 16 %P 25+ %K AI05 H01 Dragon Systems Kurzweil Products %X IBM's speech recognition system can recognize utterances in real time from a 5000 word pre-programmed vocabulary and can transcribe sentences with 95 per cent accuracy. The system may become a product. It can handle office correspondence in its present form. The system requires that the user speaks slowly and with pauses. The system runs on a PC/AT with specialized speech recognizing circuits. Kurzweil Applied Intelligence has a system with a 1000 word recognition system selling for $65,000 that has been delivered to several hundred customers. They have working prototypes of systems with 5000 word vocabularies which requires only a 1/10 of a second pause. Dragon Systems has a system that can recognize up to 1000 words. %A Stephen F. Fickas %T Automating the Transformational Development of Software %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1268-1277 %K AA08 Glitter package routing %X Describes a system to automate the selection of transformations to be applied in creating a program from a specification. Goes through an example to route packages through a network consisting of binary trees. %A Douglas R. Smith %A Goirdon B. Kotik %A Stephen J. Westwold %T Research on Knowledge-Based Software Environments at Kestrel Institute %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1278-1295 %K AA08 CHI %X Describes the CHI project. REFINE, developed by Reasoning Systems Inc., is based onthe principles and ideas demonstrated in the CHI prototype. CHI has bootstrapped itself. This system is a transformation based system. The specification language, V, takes 1/5 to 1/10 the number of lines as the program being specified if it was written in LISP. %A Richard C. Waters %T The Programmer's Apprentice: A Session with KBEmacs %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1296-1320 %K AA08 Ada Lisp %X This system, which uses plans to work hand-in-hand with a programmer in constructing a piece of software is now being used to work with ADA programs. The example used is that of a simple report. Currently, KBEmacs knows only a few dozen types of plans out of a few hundred to a few thousand for real work. Some operations take five minutes, but it is expected that a speedup by a factor of 30 could be done by straightforward operations. It is currently 40,000 lines of LISP code. %A David R. Barstow %T Domain-Specifific Automatic Programming %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1321-1336 %K AA08 AA03 well-log Schlumberger-Doll %X This system describes a system to write programs to do well-log interpretation. This system contains knowledge about well-logs as well as programming. %A Robert Neches %A William R. Swartout %A Johanna D. Moore %T Enhanced Maintenance and Explanation of Expert Systems Through Explicit Models of Their Development %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1337-1350 %K AA08 AI01 %X Describes a system for applying various transformations to improve readability of a LISP program. Also discusses techniques for providing explanation of the operation of the LISP machine by looking at data structures created as the expert system is built %A Beth Adelson %A Elliot Soloway %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1351-1360 %K AA08 AI08 %X discusses protocol analysis of designers designing software systems. Tries to show the effect of previous experience in the domain on these operations %A Elaine Kant %T Understanding Automating Algorithm Design %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1361-1374 %K AA08 AI08 %X protocol analysis on algorithm designers faced with the convex hull problem. Discussion of AI programs to design algorithms. %A David M. Steier %A Elaine Kant %T The Roles of Execution and Analysis in Design %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1375-1386 %K AA08 %A J. Doyle %T Expert Systems and the Myth of Symbolic Reasoning %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1386-1390 %K AI01 O02 %X compares traditional application development software engineering approaches with those taken by the AI community %A P. A. Subrahmanyam %T The "Software Engineering" of Expert Systems: Is Prolog Appropriate? %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1391-1400 %K T02 O02 AI01 %X discusses developing expert systems in PROLOG %A Daniel G. Bobrow %T If Prolog is the Answer, What is the Question? or What it Takes to Support AI Programming Paradigms %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 1401-1408 %K T02 AI01 %T Japanese Urge Colleges to Teach Programmers %J InfoWorld %D April 14, 1986 %V 8 %N 15 %P 18 %K GA01 %X "A panel of experts at the Japanese Ministry of Education has urged that enrollment in computer software-related departments at Japanese universities and colleges be doubled by 1992. The panel hopes to ensure that more systems engineers and software specialists are trained to offset the shortage of Japanese programmers. An estimated 600,000 additional programmers will be needed by 1990, the panel projected." %T Germans Begin AI Work with $53 Million Grant %J InfoWorld %D April 14, 1986 %V 8 %N 15 %P 18 %K GA03 %K Siemens West Germany GA03 AT19 %X The Wester German government will be giving $53.8 million in grants for AI research. %T Resources %J InfoWorld %D April 14, 1986 %V 8 %N 15 %P 19 %X New newsletter: "AI capsule", costing $195 a year for 12 issues Winters Group, Suite 920 Building, 14 Franklin Street, Rochester New York 14604 %J Electronic News %V 32 %N 1603 %D MAY 26, 1986 %P 25 %K GA01 H02 T02 Mitsubishi %X Mitsubishi Electric announces an AI workstation doing 40,000 Prolog Lips costing $118.941. %T Image-Processing Module Works like a VMEBUS CPU %J Electronics %D JUN 16, 1986 %P 74 %V 59 %N 24 %K AI06 AT02 Datacube VMEbus Analog Devices %X Product Announcement: VMEbus CPU card containing a digital signal-processing chip supporting 8 MIPS %T Robot Info Automatically %J IEEE Spectrum %D JAN 1986 %V 23 %N 1 %P 96 %K AT09 AT02 AI07 %X Robotics database available on diskette of articles on robots. Cost $90.00 per year, "Robotics Database, PO BOX 3004-17 Corvallis, Ore 97339 %A John A. Adams %T Aerospace and Military %J IEEE Spectrum %D JAN 1986 %V 23 %N 1 %P 76-81 %K AA19 AI06 AI07 AA18 AI01 %X Darpa's Autonomous Land Vehicle succeeded in guiding itself at 5 kilometers per hour using a vision system along a paved road. %A Richard L. Henneman %A William B. Rouse %T On Measuring the Complexity of Monitoring and Controlling Large-Scale Systems %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-16 %N 2 %D March/April 1986 %P 193-207 %K AI08 AA20 %X discusses the effect of number of levels of hierarchy, redundancy and number of nodes on a display page on the ability of human operators to find errors in a simulated system %A G. R. Dattatreya %A L. N. Kanal %T Adaptive Pattern Recognition with Random Costs and Its Applications to Decision Trees %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 208-218 %K AI06 AA01 AI04 AI01 clustering spina bifida bladder radiology %X applies clustering algorithm to results of reading radiographs of the bladder. The system was able to determine clusters that corresponded to those of patients with spina bifida. %A Klaus-Peter Adlassnig %T Fuzzy Set Theory in Medical Diagnosis %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 260-265 %K AA01 AI01 O04 %X They developed systems for diagnosing rheumatologic diseases and pancreatic disorders. They achieved 94.5 and 100 percent accuracy, respectively. %A William E. Pracht %T GISMO: A Visual PRoblem Structuring and Knowledge-Organization Tool %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 265-270 %K AI13 AI08 Witkin Geft AA06 %X discusses the use of a system for displaying effect diagrams on decision making in a simulated business environment. The tool improved net income production. The tool provided more assistance to those who were more analytical than to those who used heuristic reasoning as measured by the Witkin GEFT. %A Henri Farreny %A Henri Prade %T Default and Inexact Reasoning with Possiblity Degrees %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 270-276 %K O04 AI01 AA06 %X discusses storing for each proposition, a pair consisting of the probability that it is true and probability that it is false where these two probabilities do not necessarily add up to 1. Inference rules have been developed for such a system including analogs to modus ponens, modus tollens and how to combine two such ordered pairs applying to the same fact. These have been applied to an expert system in financial analysis. %A Chelsea C. White, III %A Edward A. Sykes %T A User Preference Guided Approach to Conflict Resolution in Rule-Based Expert Systems %J IEEE Transactions on Software Engineering %D NOV 1985 %V SE-11 %N 11 %P 276-278 %K AI01 multiattribute utility theory %X discusses an application of multiattribute utility theory to resolve conflicts between rules in an expert system. %A David Bright %T Chip Triggers Software Race %J ComputerWorld %V 20 %N 30 %D JUL 28, 1986 %P 1+ %K intel 3086 T01 T03 H01 Gold Hill Computers Arity Lucid T02 Hummingbird Franz %X Gold HIll Computers, Franz, Arity, Lucid, Quintus and Teknowledge have agreed to port their AI software to the 80386 %A David Bright %T Voice-activated Writer's Block %J ComputerWorld %V 20 %N 30 %D JUL 28, 1986 %P 23+ %K AI05 Kurzweill Victor Zue %X MIT's Victor Zue says that current voice recognition technology is not ready to be extended to "complex tasks." They have been able to train researchers to transcribe unknown sentences from spectrograms with 85% success. A Votan Survey showed that 87% of office workers require only 45 words to run their typical applications. Votan's add-in boards can recognized 150 words at a time. %A David Bright %T Nestor Software Translates Handwriting to ASCII code %J ComputerWorld %V 20 %N 30 %D JUL 28, 1986 %P 23+ %K AI06 Brown University %X Nestor has commercial software that converts handwriting entereded via a digitizing tablet into ascii text. First user: a French insurance firm. The system has been trained to recognize Japanese kanji characters and they will develop a video system to read handwritten checks. %A Namir Clement Shammas %T Turbo Prolog %J Byte %D SEP 1986 %V 11 %N 9 %P 293-295 %K T02 H01 AT17 %X another review of Turbo-Prolog %A Bruce Webster %T Two Fine Products %J Byte %D SEP 1986 %V 11 %N 9 %P 335-347 %K T02 H01 AT17 Turbo-Prolog %X yet another review of Turbo-Prolog %A Karen Sorensen %T Expert Systems Emerging as Real Tools %J Infoworld %V 8 %N 16 %P 33 %D APR 21, 1986 %K AI01 AT08 %A Rosemary Hamilton %T MVS Gets Own Expert System %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 1 %K T03 IBM %X IBM introduced expert system tools for the MVS operating system similar to those already introduced for VM. The run-time system is $25,000 per month while development environment is $35,000 per month. %A Amy D. Wohl %T On Writing Keynotes: Try Artificial Intelligence %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 17 %X tongue in cheek article about the "keynote" speech which appears at many conferences. (Not really about AI) %A Elisabeth Horwitt %T Hybrid Net Management Pending %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 19 %K AA08 AI01 AT02 Symbolics Avant-Garde nettworks AA15 H02 %X Avant-Garde Computer is developing an interface to networks to assist in the management thereof. Soon there will be an expert sytem on a Symbolics to interface to that to assist the user of the system. %T Software Notes %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 29+ %K ultrix DEC VAX AT02 T01 %X DEC has announce a supported version of VAX Lisp for Ultrix %A Jeffrey Tarter %T Master Programmers: Insights on Style from Four of the Best %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 41+ %K Jeff Gibbons O02 Palladian AA06 %X contains information on Jeff Gibbons, a programmer at Palladian which does financial expert systems %T Software and Services %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 76 %K T02 Quintus PC/RT AT02 %X Quintus has ported its Prolog to the IBM PC/RT. It costs $8000.00 %T New Products/Microcomputers %J ComputerWorld %D APR 7, 1986 %V 20 %N 14 %P 81-82 %K AT02 AI06 %X ADS has announced a real-time digitizer for use with micros costing between $15,000 and $25,000 %A David Bright %T Datacopy Presents Text, Image Scanner for IBM PC Family %J ComputerWorld %D APR 28, 1986 %V 20 %N 17 %P 36 %K H02 AT02 AI06 %X For $2950 you can get an integrated text and iamge scanner which can convert typewritten text to ASCII code. It can be trained to recognize unlimited numbers of fonts. It can also be used to input 200 x 200 or 300 x 300 dot per inch resolution images. %T Lisp to Separate Sales, Marketing %J Electronic News %P 27 %D APR 14, 1986 %V 32 %N 1597 %K H02 LMI AT11 %X Lisp Machines is separating sales and marketing. Ken Johnson, the former vice-president of sales and marketing, has left LMI for VG Systems %A Steven Bruke %T Englishlike 1-2-3 Interface Shown %J InfoWorld %D APR 28, 1986 %P 5 %V 8 %N 17 %K Lotus AI02 H01 AA15 %X Lotus is selling HAL, which allows users to access 1-2-3 using English commands %T TI Sells Japan Lisp Computer %J Electronics %D JUN 2, 1986 %P 60 %V 59 %N 22 %K GA02 GA01 H02 AT16 %X C. Itoh has agreed to market TI's Lisp Machine %A Larry Waller %T Tseng Sees Peril in Hyping of AI %J Electronics %D APR 21, 1986 %P 73 %V 59 %N 16 %K Hughes AT06 AI06 AI07 %X Interview with David Y. Tseng, head of the Exploratory Studies Department at Malibu Research Laboratories. %T Image Processor Beats 'Real Time' %J Electronics %P 54+ %D APR 14, 1986 %V 59 %N 15 %K AI06 AT02 H01 Imaging Technology %X Imaging Technology's Series 151 will process an image in 27 milliseconds and offers the user the ability to select an area to be processed. It interfaces to a PC/AT. It costs $11,495 with an optional convolution board for $3,995. %A A. P. Sage %A C. C. White, III %T Ariadne: A Knowledge Based Interactive System for Planning and Decision Support %J IEEE Transactions on Systems, Man, Cybernetics %V SMC-14 %D JAN/FEB 1984 %N 1 %P 48-54 %K AI13 %A R. M. Hunt %A W. B. Rouse %T A Fuzzy Rule-Based Model of Human Problem Solving %J IEEE Transactions on Systems, Man, Cybernetics %V SMC-14 %D JAN/FEB 1984 %N 1 %P 112-119 %K AI08 AI01 AA21 %X attempt to develop a model of how people diagnose engine performance %A I. B. Turksen %A D. D. W. Yao %T Representations of Connectives in Fuzzy Reasoning: The View Through Normal Forms %J IEEE Transactions on Systems, Man, Cybernetics %V SMC-14 %D JAN/FEB 1984 %N 1 %P 146-151 %K O04 %A W. X. Xie %A S. D. Bedrosian %T An Information Measure for Fuzzy Sets %J IEEE Transactions on Systems, Man, Cybernetics %V SMC-14 %D JAN/FEB 1984 %N 1 %P 151-157 %K O04 %A S. Miyamoto %A K. Nakayama %T Fuzzy Information Retrieval Based on a Fuzzy Pseudothesaurus %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-16 %N 2 %D MAR/APR 1986 %P 278-282 %K AA14 O04 %X A fuzzy bibliographic information retrieval based on a fuzzy thesaurus or on a fuzzy pseudothesaurus is described. A fuzzy thesaurus consists of two fuzzy relations defined on a set of keywords for the bibliography. The fuzzy relations are generated based on a fuzzy set model, which describes association of keyword to its concepts. If the set of concepts in the fuzzy set model is replaced by the set of documents, the fuzzy relations are called a pseudothesaurus, which is automatically generated by using occurrence frequencies of the keywords in the set of documents. The fuzzy retrieval uses two fuzzy relations in addition, that is, a fuzzy indexing and a fuzzy inverted file: the latter is the inverse relation of the former. They are, however, related to different algorithms for indexing and retrieval, respectively. An algorithm of ordering retrieved documents according to the values of the fuzzy thesaurus is proposed. This method of the ordering is optimal in the sense that one can obtain documents of maximum relevance in a fixed time interval. ------------------------------ End of AIList Digest ******************** From csnet_gateway Fri Oct 10 02:46:49 1986 Date: Fri, 10 Oct 86 02:46:34 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #209 Status: R AIList Digest Thursday, 9 Oct 1986 Volume 4 : Issue 209 Today's Topics: Bibliographies - Correction and Future SMU Bibliography Labels & Recent Kansas Technical Reports & UCLA Technical Reports ---------------------------------------------------------------------- Date: WED, 10 JAN 84 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Correction and Future SMU Bibliography Labels [Lawrence Leff at SMU, who provides all those lengthy bibliographies and article summaries, has sent the following correction for the Subject line I added to one of the bibliographies. -- KIL] ai.bib35 was mistitled as references on computer vision/robotics. This reference list contained articles on such subjects as neural networks, urban planning, logic programming, and theorem proving as well as vision/robotics. In order to prevent this problem in the future, I will entitling the materials as ai.bibnnxx where nn is a consecutive number and xx is C for citations without descriptions TR for technical reports AB for citations for citations with descriptions (annotated bibliographies) Thus ai.bib40C means the 40th AI list in bibliography format and the C indicates that we have a bunch of bib format references without significant commentary. The nn is unique over all types of bibliographies. Thus, if there were an ai.bib40C, then there will NOT be an ai.bib40TR or ai.bib40AB. These designations are actually the file names for the list on my hard disk. The shell script that wraps up the item for mailing will automatically put the file name in the subject field. If one of your readers uses this to designate a file in mail to me, I can thus trivially match their query against a specific file. Note that I no longer will be separating out references by subject matter. The keyword system is much more effective for allowing people interested in specific subfields of ai to see the articles they find relevant. Sadly the bib system program "listrefs" is having problems with citations that contain long abstracts or commentary information. Thus TR and AB type references will probably cause this program to spec check. I spent a whole day trying to isolate the problem but have been unsuccessful. One other self-described bib expert has the same problem. All references are indexable by "invert". TR and AB type references will not use bib definition files and thus are usable with the refer package from AT&T. If I were not to use bib definition files with C type reference lists, the number of bytes transmitted for their mailing would triple. ------------------------------ Date: Fri, 5 Sep 86 15:05:40 CDT From: Glenn Veach Subject: Recent Kansas Technical Reports Following is a list of technical reports which have recently been issued by the department of Computer Science of The University of Kansas in conjunction with research done in the department's Artificial Intelligence Laboratory. Requests for any and all Technical Reports from the Department of Computer Science and it's various laboratories at The University of Kansas should be sent to the following address: Linda Decelles, Office Manager 110 Strong Hall Department of Computer Science The University of Kansas Lawrence, KS 66045 U.S.A. %A Glenn O. Veach %T The Belief of Knowledge: Preliminary Report %I Department of Computer Science, The University of Kansas %R TR-86-15 %X As various researchers have attempted to present logics which capture epistemic concepts they have encountered several difficulties. After surveying the critiques of past efforts we propose a logic which avoids these same faults. We also closely explore fundamental issues involved in representing knowledge in ideal and rational agents and show how the similarities and differences are preserved in the logic we present. Several examples are given as supporting evidence for our conclusions. To be published in the proceedings of the 2nd Kansas Conference: Knowledge-Based Software Development. 12 pp. %A Glenn O. Veach %T An Annotated Bibliography of Systems and Theory for Distributed Artificial Intelligence %I Department of Computer Science, The University of Kansas %R TR-86-16 %X This paper summarizes, with extensive comment, the results of an initial investigation of the work in distributed AI. Some forty-plus articles representing the major schools of thought and development are cited and commented upon. %A Frank M. Brown %T Semantical Systems for Intensional Logics Based on the Modal Logic S5+Leib %I Department of Computer Science, The University of Kansas %R TR-86-17 %X This paper contains two new results. First it describes how semantical systems for intensional logics can be represented in the particular modal logic which captures the notion of logical truth. In particular, Kripke semantics is developed from this modal logic. The second result is the development in the modal logic of a new semantical system for intensional logics called B-semantics. B-semantics is compared to Kripke semantics and it is suggested that it is a better system in a number of ways. ------------------------------ Date: Tue, 7 Oct 86 13:32:32 PDT From: Judea Pearl Subject: new Technical Reports The following technical reports are now available from the Cognitive Systems Laboratory Room 4712, Boelter Hall University of California Los-Angeles, CA, 90024 or: judea@locus.ucla.edu _______ Pearl, J., ``Bayes and Markov Networks: a Comparison of Two Graphical Representations of Probabilistic Knowledge,'' Cognitive Systems Laboratory Technical Report (R-46), September 1986. ABSTRACT This paper deals with the task of configuring effective graphical representation for intervariable dependencies which are embedded in a probabilistic model. It first uncovers the axiomatic basis for the probabilistic relation `` x is independent of y , given z ,'' and offers it as a formal definition for the qualitative notion of informational dependency. Given an initial set of such independence relationships, the axioms established permit us to infer new independencies by non-numeric, logical manipulations. Using this axiomatic basis, the paper determines those properties of probabilistic models that can be captured by graphical representations and compares the characteristics of two such representations, Markov Networks and Bayes Networks. A Markov network is an undirected graph where the links represent symmetrical probabilistic dependencies, while a Bayes network is a directed acyclic graph where the arrows represent causal influences or object-property relationships. For each of these two network types, we establish: 1) a formal semantic of the dependencies portrayed by the networks, 2) an axiomatic characterization of the class of dependencies capturable by the network, 3) a method of constructing the network from either hard data or expert judgments and 4) a summary of properties relevant to its use as a knowledge representation scheme in inference systems. _______ Zukerman, I. & Pearl, J., ``Comprehension-Driven Generation of Meta-Technical Utterances in Math Tutoring,'' UCLA Computer Science Department Technical Report CSD-860097 (R-61). ABSTRACT A technical discussion often contains conversational expressions like ``however,'' ``as I have stated before,'' ``next,'' etc. These expressions, denoted Meta-technical Utterances (MTUs) carry important information which the listener uses to speed up the comprehension process. In this research we model the meaning of MTUs in terms of their anticipated effect on the listener comprehension, and use these predictions to select MTUs and weave them into a computer generated discourse. This paradigm was implemented in a system called FIGMENT, which generates commentaries on the solution of algebraic equations. _______ Pearl, J., ``Jeffrey's Rule and the Problem of Autonomous Inference Agents,'' UCLA Cognitive Systems Laboratory Technical Report (R-62), June 1986, UCLA CSD #860099, June 1986. ABSTRACT Jeffrey's rule of belief revision was devised by philosophers to replace Bayes conditioning in cases where the evidence cannot be articulated propositionally. This paper shows that unqualified application of this rule often leads to paradoxical conclusions, and that to determine whether or not the rule is valid in any specific case, one must first have topological knowledge about one's belief structure. However, if such topological knowledge is, indeed, available, belief updating can be done by traditional Bayes conditioning; thus, arises the question of whether it is ever necessary to use Jeffrey's rule in formalizing belief revision. _______ Pearl, J., ``Distributed Revision of Belief Commitment in Multi- Hypotheses Interpretation,'' UCLA Computer Science Department Technical Report CSD-860045 (R-64), June 1986; presented at the 2nd AAAI Workshop on Uncertainty in Artificial Intelligence, Philadelphia, PA., August 1986. ABSTRACT This paper extends the applications of belief-networks models to include the revision of belief commitments, i.e., the categorical instantiation of a subset of hypotheses which constitute the most satisfactory explanation of the evidence at hand. We show that, in singly-connected networks, the most satisfactory explanation can be found in linear time by a message-passing algorithm similar to the one used in belief updating. In multiply- connected networks, the problem may be exponentially hard but, if the network is sparse, topological considerations can be used to render the interpretation task tractable. In general, finding the most probable combination of hypotheses is no more complex than computing the degree of belief for any individual hypothesis. _______ Geffner, H. & Pearl, J., ``A Distributed Approach to Diagnosis,'' UCLA Cognitive Systems Laboratory Technical Report (R-66), October 1986; ABSTRACT The paper describes a distributed scheme for finding the most likely diagnosis of systems with multiple faults. The scheme uses the independencies embedded in a system to decompose the task of finding a best overall interpretation into smaller sub- tasks of finding the best interpretations for subparts of the net, then combining them together. This decomposition yields a globally-optimum diagnosis by local and concurrent computations using a message-passing algorithm. The proposed scheme offers a drastic reduction in complexity compared with other methods: attaining linear time in singly-connected networks and, at worst, exp ( | cycle-cutset | ) time in multiply-connected networks. _______ Pearl, J., ``Evidential Reasoning Using Stochastic Simulation of Causal Models,'' UCLA Cognitive Systems Laboratory Technical Report (R-68-I), October 1986. ABSTRACT Stochastic simulation is a method of computing probabilities by recording the fraction of time that events occur in a random series of scenarios generated from some causal model. This paper presents an efficient, concurrent method of conducting the simulation which guarantees that all generated scenarios will be consistent with the observed data. It is shown that the simulation can be performed by purely local computations, involving products of parameters given with the initial specification of the model. Thus, the method proposed renders stochastic simulation a powerful technique of coherent inferencing, especially suited for tasks involving complex, non-decomposable models where ``ballpark'' estimates of probabilities will suffice. _______ Pearl, J., ``Legitimizing Causal Reasoning in Default Logics'' (note), UCLA Cognitive Systems Laboratory Technical Report (R- 69), September 1986. ABSTRACT The purpose of this note is to draw attention to certain aspects of causal reasoning which are pervasive in ordinary discourse yet, based on the author's scan of the literature, have not received due treatment by logical formalisms of common-sense reasoning. In a nutshell, it appears that almost every default rule falls into one of two categories: expectation-evoking or explanation-evoking. The former describes association among events in the outside world (e.g., Fire is typically accompanied by smoke.); the latter describes how we reason about the world (e.g., Smoke normally suggests fire.). This distinction is clearly and reliably recognized by all people and serves as an indispensible tool for controlling the invocation of new default rules. This note questions the ability of formal systems to reflect common-sense inferences without acknowledging such distinction and outlines a way in which the flow of causation can be summoned within the formal framework of default logic. _______ Dechter, R. & Pearl, J., ``The Cycle-Cutset Method for Improving Search Performance in AI Applications,'' UCLA Cognitive Systems Laboratory Technical Report (R-67); submitted to the 3rd IEEE Conference on Artificial Intelligence Applications. ABSTRACT This paper introduces a new way of improving search performance by exploiting an efficient method available for solving tree-structured problems. The scheme is based on the following observation: If, in the course of a backtrack search, we remove from the constraint-graph the nodes corresponding to instantiated variables and find that the remaining subgraph is a tree, then the rest of the search can be completed in linear time. Thus, rather than continue the search blindly, we invoke a tree-searching algorithm tailored to the topology of the remaining subproblem. The paper presents this method in detail and evaluates its merit both theoretically and experimentally. ------------------------------ End of AIList Digest ******************** From csnet_gateway Fri Oct 10 02:48:06 1986 Date: Fri, 10 Oct 86 02:47:53 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe Subject: AIList Digest V4 #210 Status: R AIList Digest Thursday, 9 Oct 1986 Volume 4 : Issue 210 Today's Topics: Conferences - Expert Systems in Government & IEEE Systems, Man and Cybernetics ---------------------------------------------------------------------- Date: Wed, 01 Oct 86 13:03:52 -0500 From: Duke Briscoe Subject: Final Program for Expert Systems in Government Conference The Second Annual Expert Systems in Government Conference, sponsored by the Mitre Corporation and the IEEE Computer Society in association with the AIAA National Capital Section will be held October 20-24, 1986 at the Tyson's Westpark Hotel in McLean, VA. There is still time to register, but late registration charges will be added after October 6. October 20-21 Tutorials Monday, October 20 Full Day Tutorial: Advanced Topics in Expert Systems by Kamran Parsaye, IntelligenceWare, Inc. Morning Tutorial: Knowledge Base Design for Rule Based Expert Systems by Casimir Kulikowski, Rutgers University Afternoon Tutorial: Knowledge Base Acquisition and Refinement by Casimir Kulikowski, Rutgers University Tuesday, October 21 Morning Tutorial: Distributed Artificial Intelligence by Barry Silverman, George Washington University Morning Tutorial: Introduction to Common Lisp by Roy Harkow, Gold Hill Afternoon Tutorial: Lisp for Advanced Users by Roy Harkow, Gold Hill Afternoon Tutorial: The Management of Expert System Development by Nancy Martin, Softpert Systems October 22-24 Technical Program Wednesday, October 22 9 - 10:30 Conference Chairman's Welcome Keynote Address: Douglas Lenat, MCC Program Agenda 11am - 12pm Track A: Military Applications I K. Michels, J. Burger; Missile and Space Mission Determination Major R. Bahnij, Major S. Cross; A Fighter Pilot's Intelligent Aide for Tactical Mission Planning Track B: Systems Engineering R. Entner, D. Tosh; Expert Systems Architecture for Battle Management H. Hertz; An Attribute Referenced Production System B. Silverman; Facility Advisor: A Distributed Expert System Testbed for Spacecraft Ground Facilities 12pm - 1pm Lunch, Distinguished Guest Address, Harry Pople, University of Pittsburgh 1pm - 2:30pm Track A: Knowledge Acquisition G. Loberg, G. Powell Acquiring Expertise in Operational Planning: A Beginning J. Boose, J. Bradshaw; NeoETS: Capturing Expert System Knowledge K. Kitto, J. Boose; Heuristics for Expertise Transfer M. Chignell; The Use of Ranking and Scaling in Knowledge Acquisition Track B: Expert Systems in the Nuclear Industry D. Sebo et al.; An Expert System for USNRC Emergency Response D. Corsberg; An Object-Oriented Alarm Filtering System J. Jenkins, W. Nelson; Expert Systems and Accident Management 3pm - 5pm Track A: Expert Systems Applications I W. Vera, R. Bolczac; AI Techniques Applied to Claims Processing R. Tong, et al.; An Object-Oriented System for Information Retrieval D. Niyogi, S. Srihari; A Knowledge-based System for Document Understanding R. France, E. Fox; Knowledge Representation in Coder Track B: Diagnosis and Fault Analysis M. Taie, S. Srihari; Device Modeling for Fault Diagnosis Z. Xiang, S. Srihari; Diagnosis Using Multi-level Reasoning B. Dixon; A Lisp-Based Fault Tree Development Environment Panel Track: 1pm - 5pm Management of Uncertainty in Expert Systems Chair: Ronald Yager, IONA College Participants: Lofte Zadeh, UC Berkeley Piero Bonnisone, G.E. Laveen Kanal, University of Maryland Peter Cheeseman, NASA-Ames Research Center Prakash Shenoy, University of Kansas Thursday, October 23 9am - 10:30am Track A: Knowledge Acquistion and Applications E. Tello; DIPOLE - An Integrated AI Architecture H. Chung; Experimental Evaluation of Knowledge Acquisition Methods H. Gabler; IGOR - An Expert System for Crash Trauma Assessment K. Chhabra, K. Karna; Expert Systems in Electronic Filings Track B: Aerospace Applications of Expert Systems D. Zoch; A Real-time Production System for Telemetry Analysis J. Schuetzle; A Mission Operations Planning Assistant D. Brauer, P. Roach; Ada Knowledge Based Systems F. Rook, T. Rubin; An Expert System for Conducting a Sattelite Stationkeeping Maneuver Panel Track: Star Wars and AI Chair: John Quilty, Mitre Corp. Participants: Brian P. McCune, Advanced Decision Systems Lance A. Miller, IBM Edward C. Taylor, TRW 11am - 12pm Plenary Address: B. Chandrasekaran; The Future of Knowledge Acquisition 12pm - 1pm Lunch 1pm - 2:30pm Track A: Inexact and Statistical Measures K. Lecot; Logic Programs with Uncertainties N. Lee; Fuzzy Inference Engines in Prolog/P-Shell J. Blumberg; Statistical Entropy as a Measure of Diagnostic Uncertainty Track B: High Level Tools for Expert Systems S. Shum, J.Davis; Use of CSRL for Diagnostic Expert Systems E. Dudzinski, J. Brink; CSRL: From Laboratory to Industry D. Herman, J. Josephson, R. Hartung; Use of the DSPL for the Design of a Mission Planning Assistant J. Josephson, B. Punch, M. Tanner; PEIRCE: Design Considerations for a Tool for Abductive Assembly for Best Explanation Panel Track: Application of AI in Telecommunications Chair: Shri Goyal, GTE Labs Participants: Susan Conary, Clarkson University Richard Gilbert, IBM Watson Research Center Raymond Hanson, Telenet Communications Edward Walker, BBN Richard Wolfe, ATT Bell Labs 3pm - 5pm Track A: Expert System Implementations S. Post; Simultaneous Evaluation of Rules to Find Most Likely Solutions L. Fu; An Implementation of an Expert System that Learns R. Frail, R. Freedman; OPGEN Revisited R. Ahad, A. Basu; Explanation in an Expert System Track B: Expert System Applications II R. Holt; An Expert System for Finite Element Modeling A. Courtemanche; A Rule-based System for Sonar Data Analysis F. Merrem; A Weather Forecasting Expert System Panel Track: Command and Control Expert Systems Chair: Andrew Sage, George Mason University Participants: Peter Bonasso, Mitre Stephen Andriole, International Information Systems Paul Lehner, PAR Leonard Adelman, Government Systems Corporation Walter Beam, George Mason University Jude Franklin, PRC Friday, October 24 9am - 12pm: Expert Systems in the Classified Community The community building expert systems for classified applications is unsure of the value and feasibility of some form of communication within the community. This will be a session consisting of discussions and working sessions, as appropriate, to explore these issues in some depth for the first time, and to make recommendations for future directions for the classified community. 9am - 10:30am Track A: Military Applications Bonasso, Benoit, et al.; An Experiment in Cooperating Expert Systems for Command and Control J. Baylog; An Intelligent System for Underwater Tracking J. Neal et al.; An Expert Advisor on Tactical Support Jammer Configuration Track B: Expert Systems in the Software Lifecycle D. Rolston; An Expert System for Reducing Software Maintenance Costs M. Rousseau, M. Kutzik; A Software Acquisition Consultant R. Hobbs, P. Gorman; Extraction of Data System Requirements Panel Track: Next Generation Expert System Shells Chair: Art Murray, George Washington University Participants: Joseph Fox, Software A&E Barry Silverman, George Washington University Chuck Williams, Inference John Lewis, Martin Marietta Research Labs 11am - 12pm Track A: Spacecraft Applications D. Rosenthal; Transformation of Scientific Objectives into Spacecraft Activities M. Hamilton et al.; A Spacecraft Control Anomaly Resolution Expert System Track B: Parallel Architectures L. Sokol, D. Briscoe; Object-Oriented Simulation on a Shared Memory Parallel Architecture J. Gilmer; Parallelism Issues in the CORBAN C2I Representation Panel Track: Government Funding of Expert Systems Chair: Commander Allen Sears, DARPA Participants: Randall Shumaker, and others Conference Chairman: Kamal Karna Unclassified Program Chairman: Kamran Parsaye Classified Program Chairman: Richard Martin Panels Chairman: Barry Silverman Tutorials Chairman: Steven Oxman Registration information can be requested from Ms. Gerrie Katz IEEE Computer Society 1730 Massachusetts Ave. N.W. Washington, D.C. 20036-1903 (202) 371-0101 ------------------------------ Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Conference - IEEE Systems, Man and Cybernetics 1986 IEEE International Conference on Systems, Man and Cybernetics, AI papers October 14-17, 1986 Pierremont Plaza Hotel, Atlanta, GA 30308 Wednesday October 15 8AM - 9:40 AM On Neural-Model Based Cognitive Theory and Engineering: Introduction N. DeClaris Matrix and Convolution Models of Brain Organization in Cognition K. H. Pribram Explorations in Brain Style Computations D. E. Rumelhart Representing and Transforfming Recursive Objects in a Neural Network or "Trees Do Grow on Boltzmann Machines D. S. touretzky Competition-Based Connectionist Models of Associative Memory J. A. Reggia, S. Millheim, A. Freeman A Parallel Network that Lears to Read Aloud T. J. Sejnowski A Theory of Dialogue Structures to HElp Manage Human Computer Interaction D. L. Sanford, J. W. Roach A User Interface for a Knowledge-Based Planning and Scheduling System A. M. Mulvehill Orthonormal Decompositions in Adaptive Systems L. H. Sibul An "Evolving Frame" Approach to Learning with Application to Adaptive Navigation R. J. P. de Figueredo, K. H. Wang Approaches to Machine Learning with Genetic Algorithms J. Grefenstette, C. B. Pettey Use of Voice Recognition for Control of a Robotic Welding Workcell J. K. Watson, D. M. Todd, C. S. Jones A Knowledge Based System for the CST Diagnosis C. Hernandez, A. Alonso, Z. Wu A Qualitative Model of Human Interaction with Complex Dynamic Systems R. A. Hess Evaluating Natural Language Interfaces to Expert Systems R. M. Weischedel Expert System Metrics S. Kaisler Global Issues in Evaluation of Expert Systems N. E. Lane A Scenario-Based Test Tool for Examining Expert Systems E. T. Scambos 10:AM - 11:40AM A Comparison of Some Inductive Learning Methodologies D. W. Patterson INduction of Finite Automata by Genetic Algorithms H. Y. Zhou, J. J. Grefenstette NUNS: A Machine Intelligence Concept for Learning Object Class Domains B. V. Dasarathy Toward a Paradigm for Automated Understanding of 3-D Medical Images E. M.Stokely, T. L. Faber Development of an Expert System for Interpreting Medical Images N. F. Ezquerra, E. V. Garcia, E. G. DePuey, W. L. Robbins Edge Enhancement in Medical Images by 3D Processing J. E. Boyd, R. A. STein Scheme for Three Dimensional Reconstruction of Surfaces from CT and MRI Images of the Human Body R. Tello, R. W. Mann, D. Rowell Using the Walsh-Hadamard Phase Spectrum to Generate Cardiac Activation Movies- A Feasibility Study H. Li Things We Learned by Making Expert Systems to Give Installation Advice for UNIX 4.2BSD and to HElp Connect a Terminal to a Computer A. T. Bahill, E. Senn, P. Harris A Heuristic Search/Information Theory Approach to Near Optimal Diagnostic Test Sequencing K. R. Pattipati, J. C. Deckert, M. G. Alexandris An Expert System for the Estimation of Parameter Values of Water Quality MOdel W. J. Chen Application of an Expert System to Error Detection and Correction in a Speech Recognition System K. H. Loken-Kim, M. Joost, E. Fisher 1PM - 1:50PM Topic: Holonomic Brain Theory and The Concept of Information Karl H. Pribram 2-3:40PM An Interactive Machine Intelligence Development System for Generalized 2-D Shapes Recognition B. V. Dasarathy Modelling of Skilled Behaviour and Learning T. Bosser Design of a User INterface for Automated Knowledge Acquisition A. S. Wolff, B. L. Hutchins, E. L. Cochran, J. R. Allard, P. J. Ludlow OFMspert: An Operator Function Model Expert System C. M. Mitchell, K. S. Rubin, T. Govindaraj An Adaptive Medical Information System N. DeClaris Intermediate Level Heuristics for Road-finding Algorithms S. Vasudevan, R. L. cannon, J. C. Bezdek, W. C. Cameron Computer-Disambiguation of Multi-Character Key Text Entry: An Adaptive Design Approach S. H. Levine, S. Minneman, C. Getschow, C. Goodenough-Trepaigner An INteractive and Data Adaptive Spectrum Analysis System C. H. Chen, A. H. Costa On How Two-Action Ergodic Learning Automata can Utilize Apriori Information B. J. Oommen VLSI Implementation of an Iterative Image Restoration Algorithm A. K. Katsaggelos, P. R. Kumar, M. Samanthan Development of Automated Health Testing and Services System via Fuzzy Reasoning E. Tazaki, Y. Hayashi, K. Yoshida, A. Koiwa Knowledge-Based Interaction Tools R. Neches Bibliographic Information Retrieval Systems: Using Ai Techniques to Improve Cognitive Compatibility and Performance P. J. Smith, D. A. Krawczak, S. J. Shute, M. H. Chignell, M. Sater 4PM-5:40PM An Evidential Approach to Robot Sensory Fusion J. H. Graham Retinal Ganglion Cell Processing of Spatial Information in Cats J. Troy, J. G. Robson, C. Enroth--Cugel Texture Discriminants from Spatial Frequency Channels G. A. Wright, M. E. Jernigan Contextual Filters for Image Processing G. F. McLean, M. E. Jernigan Using Cognitive Psychology Techniques for Knowledge Acquisition A. H. Silva, D. C. Regan Transfer of kNowledge from Domain Expert to Expert System: Experience Gained form JAMEX J. G. Neal, D. J. Funke Metholdological Tools for kNowledg eAcquisition K. L. Kessel Downloading the Expert: Efficient Knowledge Acquisition for Expert Systems J. H. Lind Integration of Phenomenological and Fundamental Knowledge in Diagnostic Expert Systems L. M. Fu Integrating Knowledge Acquisitions Methods P. Dey, K. D. Reilly Multi Processing of Logic Programs G. J. Li, B. W. Wah A Model for Parallel Processing of Production Systems D. I. MOldovan Several IMplementaitons of Prolog, the Microarchitecture Perspective Y. N. Patt A Parallel Symbol-Matching Co-procesor for Rule Processing Sytems D. F. Newport, G. T. Alley, W. L. Bryan, R. O. Eason, D. W. Bouldin The Connection Machine Architecture W. D. Hillis, B. S. Kahle Thursday, October 16th 8AM-9:40PM Transformation Invariance Using HIgh ORder Correlations in Neural Net Architectures T. P. Maxwell, C. L. Giles, Y. C. Lee, H. H. Chen A Neural Network Digit Recognizer D. J. Burr Computational Properties of a Neural Net with a Triangular Lattice Structure and a Traveling Activity Peak R. Eckmiller Fuzzy Multiobjective Mathematical Programming's Application to Cost Benefit Analysis L. Xu Evaluation of the Cause Diagnosis Function of a Prototype Fuzzy-Logic-Based Knowledge System for Financial Ratio Analysis Analysis F. J. Ganoe, T. H. Whalen, C. D. Tabor Knowledge INtegration inFinancial Expert Systems P. D. Crigler, P. Dey Pyramid and Quadtree ARchitectures in Point Pattern Segmentation and Boundary Extraction B. G. Mobasseri Causality in Pattern Recognition S. Vishnubhatla Network Biovisitrons for HIgh-Level Pattern Recognition D. M. Clark, F. Vaziri Giving Advice as Extemporaneous Elaboration M. A. Bienkowski Dynamics of Man-Machine Interaction in a Conversational Advisory System A. V. Gershman, T. Wolf 10AM -11:40AM A Method for Medial Line Transformation E. Salari An Alternative IMplematnationStrategy for a Variety of Image Processing Algorithms R. Saper, M. E. Jernigan A Semantic Approach to Image Segmentation S. Basu A SkeletonizingAlgorithm with Improved Isotropy D. J. Healy The Application of Artificial INtelligence to Manufacturing control P. J. O'Grady, K. H. Lee, M. Brightman An Expert System for designof Flexible Manufacturing Systems D. E. Brown, G. anandalingam A Derivational Approach to Plan Refinement for Advice Giving R. Turner TheRole of Plan Recognition in Design of an INtelligent User INterface C. A. Broverman, K. E. Khuff, V. Lesser Discussant J. L. Koldner Voice INput in Real time Decision Making M. G. Forren, C. M. Mitchell 2:00PM-3:40PM The Use of Artificial Intelligence in CAI for Science Education G. S. Owen Design of an Intelligent Tutoring System (ITS) for Aircraft Recognition D. R. Polwell, A. E. Andrews A Rule-Based Bayesian ARchitecture for Monitoring Learnign Process in ICAI Systems T. R. Sivasankaran T. Bui A Knowledge Based System for Transit Planning A. Mallick, A. Boularas, F. DiCesare On the Acquisition and Processingo f Uncertian Information in Rule-Based Decision Support Systems S. Gaglio, R. Minciardi, P. P. Puliafito Lambertian Spheres Parameter Estimation from a Single 2-D Image B. Cernuschi-Frias D. B. Cooper A Solution to the STereo Correspondence Problem using Disparity Smoothness Constraints N. H. Kim, A. C. Bovik Registration of Serial Sectional Images for 3-D Reconstruction M. Sun, C. C. li Rotation-Invariant contour DP Matching Method for 3D Object Recognition H. Yamada, M. Hospital, T. Kasvand CAD Based 3-D Models for Computer Vision B. Bhanu, C. C. Ho, S. Lee A Rule-Based System for Forming Sequence Design for Multistage Cold Forging K. Sevenler, T. Altan, P. S. Raghupathi, R. A. Miller Automated Forging Design A. Tang Geometry Representation to Aid Autoamted Design on Blocker Forging K. R. Vemuri Intelligent Computing ("The Sixth Generation"): A Japanese Initiative R. E. Chapman The Influenceof the United States and Japan on Knowledge Systems of the Future B. A. Galler Knowledgeis Structured in Conscioiusness T. N. Scott, D. D. Scott Knowledge Science-Towards the Prosthetic Brain M. L. Shaw Socio-Economic Foundations of Knowledge Science B. R. Gaines 4:00 PM - 5:40PM Fuzzy and Vector Measurement of Workload N. Moray, P. Eisen, G. Greco, E.Krushelnycky, L. Money, B. Muir, I. Noy, F. Shein, B. Turksen, L. Waldon Toward an Empirically-based Process Model for a Machine Programming Tutor D. Littman, E. Soloway An Intelligent Tutor for Thinking about Programming J. Bonar An Expert System for Partitioning and Allocating Algorithms M. M. Jamali, G. A. Julien, S. L. Ahmad A Knowledge INcreasing Model of Image Understanding G. Tascini, P. Puliti An Artificial Intelligence Approach for Robot-Vision in Assembly Applications Environment K. Ouriachi, M. Bourton Visible Surface Reconstruction under a Minimax Criterion C. Chu, A. C. Bovak A Measurement of Image Concordance Using Replacment Rules R. Lauzzana High-Level Vision Using a Rule-Based Language M. Conlin An Expert Consultant for Manufacturing Process Selection A. Kar A Knowledge Representation Scheme for Processes in an Automated Manufacturing Environment S. R. Ray Making Scheduling Desisions in an F. M. S. Using the State-Operator Framework in A. I. S. de, A. Lee Intelligent Exception Processing for Manufacturing Workstation Control F. DiCesare, A. Desrochers, G. Goldbergen Knowledge of Knowledge and the Comptuer J. A. Wojciechowski Paradigm Chagne in the Sixth Generation Approach W. H. C. Simmonds Educational Implications of Knowledge Science P. Zorkoczy >From Brain Theory to the Sixth Generation Computer M. A. Arbib Friday, October 17 8:00 AM - 9:40 AM Development of an Intelligent Tutoring System K. Kawamura, J. R. Bourne, C. Kinzer, L. Cozean, N. Myasaka, M. Inui CALEB: An Intelligent Second Language Tutor P. Cunningham, T. Iberall, B. Woolf A Methodology for Development of a Computer-Aided Instruction Program in Complex, Dynamic Systems J. L. Fath, C. M. MItchell, T. Govindaraj Matching Strategies in Error Diagnosis: A Statistics Tutoring Aid M. M. Sebrechts, L. J. Schooler, L. LaClaire Using Prolog for Signal Flow Graph Reduction C. P. Jobling, P. Grant A Self-Organizing Soft Clustering Algorithm M. A. Ismail A Modified Fisher Criterion for Feature Extraction A. Atiya A Model of Human Kanji Character Recognition K. Yokosawa, M. Umeda, E. Yodogawa Efficient Recognition of Omni-Font Characters using Models of Human Pattern Perception D. A. Kerrick, A. C. Bovik Printed Character Recognition Using an Artificial Visual System J. M. Coggins, J. T. Poole Multiobjective INtelligent Computer Aided Design E. A. Sykes, C. C. White Knowledge Engineering for Interactive Tactical Planning: A Tested Approach with General Purpose Potential S. J. Andriole ESP- A Knowledge-Aided Design Tool J. F. King, E. Hushebeck A Study of Expert Decision Making in Design Processes R. M. Cohen, J. H. May, H. E. Pople An Intelligent Design Aid for Large Scale Systems with Quantity Discount Pricing A. R. Spillane, D. E. Brown 10AM - 1140AM NeoETS: Interactive Expertise Transfer for Knowledge-Based Systems J. H. Boose, J. M. Bradshaw PCS: A Knowledge-Based Interactive System for Group Problem Solving M. L. Shaw Cognitive Models of Human-Computer INteraction in Distributed Systems B. R. Gaines The Use of Expert Systems to Reduce Software Specification Errors S. B. Ahmed, K. Reside Structure Analysis for Gray Level Pictures on a Mesh Connected Computer J. El Mesbahi, J. S. Cherkaoui Pattern Classification on the Cartesian Join System: A General Tool for Featue Selection M. Ichino Texture Discrimination using a Model of the Visual Cortex M. Clark, A. C. Bovik Surface Orientation from Texture J. M. Coggins, A. K. Jain Classificationof Surface Defects on Wood Boards A. J. Koivo, C. Kim ADEPT: An Expert System for Finite Element Modeling R. H. HoltU. Narayana KADD: An Environment for Interactive Knowledge Aided Display Design P. R. Frey, B. J. Widerholt 3PM - 3:40PM Assigning Weights and Ranking Information Importance in an Object Identification Task D. M. Allen Third Generation Expert Systems J. H. Murphy, S. C. Chay, M. M. Downs Reasoning with Comparative Uncertainty B. K. Moore On a Blackboard Architecture for an Object-Oriented Production System D. Doty, R. Wachter Pattern Analysis of N-dimensionial Digital Images E. Khalimsky ------------------------------ End of AIList Digest ******************** From csnet_gateway Fri Oct 10 04:43:21 1986 Date: Fri, 10 Oct 86 04:43:07 edt From: csnet_gateway (LAWS@SRI-STRIPE.ARPA) To: ailist@sri-stripe.arpa Subject: AIList Digest V4 #211 Status: RO AIList Digest Friday, 10 Oct 1986 Volume 4 : Issue 211 Today's Topics: Queries - Line-Drawing Recognition & Cognitive Neuroscience, Schools - Cognitive Science at SUNY, AI Tools - XILOG & Public-Domain Prolog, Review - Canadian Artificial Intelligence, Logic Programming - Prolog Multiprocessors Book, Learning - Multilayer Connectionist Learning Dissertation ---------------------------------------------------------------------- Date: 9 Oct 86 07:52:00 EDT From: "CUGINI, JOHN" Reply-to: "CUGINI, JOHN" Subject: request for references on drawings I'd appreciate getting references to any work on automatic comparison or classification of drawings, especially technical drawings and blueprints. For instance, a system which, when presented with a blueprint, can recognize it as a left-handed widget, etc. Please send replies directly to me - thanks. John Cugini ------------------------------ Date: Mon, 6 Oct 86 13:17:40 edt From: klahr@nyu-csd2.arpa (Phillip Klahr) Subject: Cognitive Neuroscience For my Neuroscience qualifying exam, I am looking for articles, books, or reviews that discuss the interface/contribution of AI research on vision and memory to "Cognitive Neuroscience". By Cognitive Neuroscience, I mean the study of theories and methods by which the different parts of the brain go about processing information, such as vision and memory. To give you an idea of "ancient works" I am starting with, I am already looking at: Wiener's "Cybernetics", von Neumann's "The Computer and the Brain", Rosenblatt's "Principles of Neurodynamics", Arbib's "Metaphorical Brain", and Hebb's "The Organization of Behavior". Some of the neurophysiology work I am looking at already includes work by Mortimer Mishkin and Larry Squire on memory in the monkey. Any pertinent references you can think of will be very much appreciated, and, if there is any interest, I will post a summary of any responses I get. Thank you very much. Phillip Klahr Albert Einstein College of Medicine klahr@NYU-CSD2.ARPA UUCP: {allegra, seismo, ihnp4} !cmcl2!csd2!klahr ------------------------------ Date: Mon, 29 Sep 86 10:46:55 EDT From: "William J. Rapaport" Subject: Re: Cognitive Science Schools In article <8609221503.AA15901@mitre.ARPA> schwamb@MITRE.ARPA writes: >Well, now that some folks have commented on the best AI schools in >the country, could we also hear about the best Cognitive Science >programs? Cog Sci has been providing a lot of fuel for thought to >the AI community and I'd like to know where one might specialize >in this. > >Thanks, Karl (schwamb@mitre) The SUNY Buffalo Graduate Group in Cognitive Science was formed to facilitate cognitive-science research at SUNY Buffalo. Its activities have focused on language-related issues and knowledge representation. These two areas are well-represented at SUNY Buffalo by the research interests of faculty and graduate students in the Group. The Group draws its membership primarily from the Departments of Computer Science, Linguistics, Philosophy, Psychology, and Communicative Disorders, with many faculty from other departments (e.g., Geography, Education) involved on a more informal basis. A current research project on deixis in narrative is being undertaken by a research subgroup. While the Group does not offer any degrees by itself, a Cognitive Science "focus" in a Ph.D. program in one of the participating disciplines is available. There is also a Graduate Group in Vision. For further details, see AI Magazine, Summer 1986, or contact: William J. Rapaport Assistant Professor of Computer Science Co-Director, Graduate Group in Cognitive Science Dept. of Computer Science, SUNY Buffalo, Buffalo, NY 14260 (716) 636-3193, 3180 uucp: ..!{allegra,decvax,watmath,rocksanne}!sunybcs!rapaport csnet: rapaport@buffalo.csnet bitnet: rapaport@sunybcs.bitnet ------------------------------ Date: 1 Oct 86 20:02:38 GMT From: mcvax!unido!ecrcvax!bruno@seismo.css.gov (Bruno Poterie) Subject: XILOG Well, I know of at least one Prolog system on PC/AT which is: - fully C&M compatible, to the exception of the top-level mode (consults by default (terms ended with a dot), executes on request (terms ended with a question mark)) - all defined C&M predicates, i/o, program manipulation, term scanning & construction, integer and float arithmetics, ... plus the following features: - full window, semi-graphics & color management - modularity for debugging, program handling, etc.. ( but *no* separate dictionaries) through a hierarchy of "zone" - on-line precise help - on-line clause editor - complete typage mechanism, allowing full object definition, constraint checking, etc... - functional mechanism, allowing each call to return any prolog term as a return value through an accumulator (bactracked/trailed) (the arithmetic is implemented, using this mechanism, and you may extend it as you wants) - non-bactrackable global cells and arrays - backtracking arrays, with functional notation and access - access to MSDOS - sound system and some others less (sic) important goodies, like a debugger based on the Box model, etc... oh, I forgot: under development, and rather advanced by now, are: - an incremental compiler to native code with incremental linking (with full integration with the interpreter, of course) - an interface to C programs - a toolkit for development of applications, with an utilities library - and maybe a message sending mechanism (but I'm not sure for it) The name of this system is: XILOG and it is made and distributed by (the Research Center of) BULL, the biggest french computer compagny. if interested, contact: CEDIAG BULL 68, route de Versailles F-78430 Louveciennes FRANCE or: Dominique Sciamma (same address) don't fear, they do speak english there! :-) P.S.: I precise that I have no commercial interest at all in this product, but I really think that this XILOG is the best Prolog for micro I ever met. ================================================================================ Bruno Poterie # ... une vie, c'est bien peu, compare' a un chat ... ECRC GmbH # tel: (49)89/92699-161 Arabellastrasse 17 # Tx: 5 216 910 D-8000 MUNICH 90 # mcvax!unido!ecrcvax!bruno West Germany # bruno%ecrcvax.UUCP@Germany.CSNET ================================================================================ ------------------------------ Date: 8 Oct 86 20:26:15 GMT From: ucdavis!ucrmath!hope!fiore@ucbvax.Berkeley.EDU (David Fiore) Subject: Re: pd prolog > Xref: ucbvax net.micro:451 net.micro.pc:821 net.ai:91 > > Does anyone have the public domain prolog package discussed in this month's > BYTE magazine? > > John E. Jacobsen > University of Wisconsin -- Madison Academic Computing Center I have a copy of pdprolog here with me. It is the educational version. I don't know if that is the one described in BYTE as I haven't read that magazine lately. || || David Fiore, University of California at Riverside. ============= || Slow mail : 1326 Wheaton Way || Riverside, Ca. 92507 || E-Mail || UseNet : ...!ucdavis!ucrmath!hope!fiore || BITNET : consult@ucrvms Have another day! "...and at warp eight, we're going nowhere mighty fast" ------------------------------ Date: WED, 20 apr 86 17:02:23 CDT From: E1AR0002%SMUVM1.BITNET@WISCVM.WISC.EDU Subject: Canadian Artificial Intelligence/ September 1986 Summary No. 9 Report on current budget and increase in dues The Dalhousie Business School got a Sperry Explorer Lisp Machine and a copy of KEE. They are developing a system to manage foreign debts and plan an estimator for R&D projects, intelligent computer aided instruction and auditing. Xerox Canada has set up an AI support work Logicware has been acquired by the Nexa Group British Columbia Advanced Systems Institute will be set up to do research on AI, robotics, microelectronics. __________________________________________________________________________ Two assessments on the Japanese Fifth Generation project: ICOT is developing AI systems for fishing fleets, train control, microchip design, natural language transition. There are 600 researchers working on fifth generation projects and 600 on robotics. 1986-1988 funding is 102 billion yen and 1982-92 funding is 288 billion. The English to Japanese system will require post-editing and applies standard techniques. The Japanese have abandoned 'Delta', their parallel inference engine is 'gathering dust' They alledgedly threw 'hardware engineers' into a Prolog environment for which they 'had no background or interest' __________________________________________________________________________ Report on Natural Language Understanding Research at University of Toronto Reviews of Bertholt Klaus Paul Horn's "Robot Vision" Theoretical Aspects of Reasoning About Knowledge: Proceedings of the 1986 Conference ------------------------------ Date: Tue, 7 Oct 86 16:08:19 EST From: munnari!nswitgould.oz!michaelw@seismo.css.gov Subject: Book - Prolog Multiprocessors A book is soon to appear, by Michael J. Wise, entitled "Prolog Multiprocessors". It is being published by Prentice-Hall (Australia). In a nutshell, the book examines the execution of Prolog on a multiprocessor. Starting from a survey of some current multiprocessor architectures, and a review of what is arguably the most influential counter-proposal - the "data-flow" model, a model is proposed for executing Prolog on a multiprocessor. Along with the model goes a language based on Prolog. The model and the language are called EPILOG. EPILOG employs both AND and OR parallelism. Results are then reported for the simulated execution of some Prolog programs rewritten in the EPILOG language. The book concludes with an extensive survey of other multiprocessor implementations of Prolog. The book will be available in Australia from mid November, and in US/UK/Europe roughly eight weeks later. A list of the Chapter headings follows. A more detailed list can be obtained from your local P-H representative, or by e-mailing to me directly. TABLE OF CONTENTS Foreword by J. Alan Robinson Preface 1. Parallel Computation and the Data-Flow Alternative 2. Informal Introduction to Prolog 3. Data-Flow Problems and a Prolog Solution 4. EPILOG Language and Model 5. Architectures for EPILOG 6. Experimenting with EPILOG Architectures - Results and Some Conclusions 7. Related Work Appendix 1 Data-Flow Research - the First Generation Appendix 2 EBNF Specification for EPILOG Appendix 3 EPILOG Test Programs Appendix 4 Table of Results ------------------------------ Date: Thu, 9 Oct 86 10:21:18 EDT From: "Charles W. Anderson" Subject: Dissertation - Multilayer Connectionist Learning The following is the abstract from my Ph.D. dissertation completed in August, 1986, at the University of Massachusetts, Amherst. Members of my committee are Andrew Barto, Michael Arbib, Paul Utgoff, and William Kilmer. I welcome all comments and questions. Chuck Anderson GTE Laboratories Inc. 40 Sylvan Road Waltham, MA 02254 617-466-4157 cwa0@gte-labs Learning and Problem Solving with Multilayer Connectionist Systems The difficulties of learning in multilayered networks of computational units has limited the use of connectionist systems in complex domains. This dissertation elucidates the issues of learning in a network's hidden units, and reviews methods for addressing these issues that have been developed through the years. Issues of learning in hidden units are shown to be analogous to learning issues for multilayer systems employing symbolic representations. Comparisons of a number of algorithms for learning in hidden units are made by applying them in a consistent manner to several tasks. Recently developed algorithms, including Rumelhart, et al.'s, error back-propagation algorithm and Barto, et al.'s, reinforcement-learning algorithms, learn the solutions to the tasks much more successfully than methods of the past. A novel algorithm is examined that combines aspects of reinforcement learning and a data-directed search for useful weights, and is shown to out perform reinforcement-learning algorithms. A connectionist framework for the learning of strategies is described which combines the error back-propagation algorithm for learning in hidden units with Sutton's AHC algorithm to learn evaluation functions and with a reinforcement-learning algorithm to learn search heuristics. The generality of this hybrid system is demonstrated through successful applications to a numerical, pole-balancing task and to the Tower of Hanoi puzzle. Features developed by the hidden units in solving these tasks are analyzed. Comparisons with other approaches to each task are made. ------------------------------ End of AIList Digest ********************