From comsat@vtcs1 Fri Dec 20 02:06:03 1985 Date: Fri, 20 Dec 85 02:05:30 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a018049; 18 Nov 85 3:42 EST Date: Sun 17 Nov 1985 23:50-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #171 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Mon, 18 Nov 85 21:58 EST AIList Digest Monday, 18 Nov 1985 Volume 3 : Issue 171 Today's Topics: Report - Limits of Correctness in Computers (CSLI), Awards - 1984 Information Awards of NSF, Review - Summary of Spang Robinson Report, Reports - ICOT Tech Reports ---------------------------------------------------------------------- Date: Wed 13 Nov 85 17:05:26-PST From: Emma Pease Subject: Report - Limits of Correctness in Computers (CSLI) Report No. CSLI-85-36, ``Limits of Correctness in Computers'' by Brian Cantwell Smith, has just been published. This report may be obtained by writing to David Brown, CSLI, Ventura Hall, Stanford, CA 94305 or Brown@SU-CSLI. ------------------------------ Date: Sun, 10 Nov 85 18:00:11 est From: Ed Fox Subject: SIGIR Forum Fall 85 - List of 1984 Information Awards of NSF [Excerpted from the IRList digest by Laws@SRI-AI.] I. INFORMATION SCIENCE PROGRAM Bolt Beranek & Newman User Goals as a Basis for an Intelligent System's Ability to Understand Ill- Formed Input--Weischedel, Ralph Brandeis Univ. Information Structure of a Natural Language Lexicon--Jackendoff, Ray Brown Univ. Cognitive Applications of Matrix Memory Models--Anderson, James A. ... II. INFORMATION TECHNOLOGY III. INFORMATION IMPACT PROGRAM [For the full 13K text of this IRList Digest, FTP file NSFIS.TXT from SRI-AI, or send a request to AIList-Request@SRI-AI.ARPA. -- KIL] ------------------------------ Date: 9 Nov 1985 12:23-CST From: leff%smu.csnet@CSNET-RELAY.ARPA Subject: Summary of Spang Robinson Report FOCUS on MICRO Expert System tool kit Estimated 1985 market at 7 million with 11,000 tool kits. Sales figures: Human Edge Software's Expert Ease 1800 copies (Jan - May) Expert Edge 426 copies (Jan - October) Expertintelligence 3000 copies Level Five Research Insight 1 and 2 1300 copies Review of Symantec's Q&A and Paradoxes by ANSI, two AI products for industry. Reviews of the following Expert System tools including a nice comparative table with prices, machines available on, units sold and revenues from expert systems. First Class: limited to 256 examples and 32 factors per knowledge base but knowledge bases can be chained. Expert Ease: same limitations as firstclass Wisdom PX: quantitative analysis Wizdom XS: problem diagnostic KDS: can handle 4000 rules and has dialog oriented user interface TIMM: will infer an answer when the case does not exactly fit any of the examples entered. Will report when case is not close enough to any example. Will generate confidence factor. Also will allow different experts to be integrated into the same expert system RuleMaster: will allow user to create rule system with induction and then edit with a rule editor. Rule Based Systems PPE public domain system available for $20.00 MicroExpert for $49.95 (originally published in Byte but updated significantly) $95.00 Insight1 from Level 5 Research Insight2 from Level5 for $495.00. Includes cyclical control and interfaces to Turbo Pascal and dBase II and dbaseIII, supports math and confidence systems Exsys: suports math and confidence factors Expert Edge: imports files, object hierarchy High End Products Personal Consultant from TI M1 from Tecknowledge KES: supports frames, production rules, Bayes Theorems and is written in C Aion: checks for detached and redundant rules, has a Think Tank-like front end. Neuron Expert: Runs on on Macintosh, allows user to print network giving relationship of all rules Expertteach, includes Prolog 86, UO-Lisp and forward and backward inferencing engines in prolog, lisp, pascal and DBASE-II VENTURE CAPITAL interviews with Sprout Group and Mayfield Fund NEWS section Applied Expert Systems agreed to buy 1000 Xerox 1186 AI workstations; Applied Expert System markets PlanPower a personal financial planning system. It will be distributing this system through Travelers Insurance system to independent financial planners MAD Intelligent systems agreed to supply compatible workstations to the Philips group of companies, expected value: 15 million Software A&E's KES now runs on the Tektronix 4404. Migent Software has released ENRICH, The MEtods Expert, $595 on the IBM PC Carnegie Group has delivered "Dispatcher" to Digital Equipment, which dispatches order and controls material handling/conveyer for a printed wire board assembly plant. Discussion of Ford deal with Inference Corporation: LIST of recent events and changes of bindings. Review of Donald A. Waterman's "Guide to Expert Systems", "Market for Expert Systems bibliography of recent articles on AI and the PC. ------------------------------ Date: Fri, 8 Nov 85 10:31:04 cst From: Xu Yan ren Subject: ICOT Tech Reports .TS box; cb s s s s c|c|c|c n|ltw(1i)|ltw(3i)|ltw(1.5). ICOT TECHNICAL REPORT _ NO. Author Topic Comment _ 002 S. Kunifuji et. T{ PROLOG and Relational Data Bases for Fifth Generation Computer system T} _ 003 E.Y. Shapiro T{ A Subset Concurrent PROLOG and Its Interpreter T} _ 006 A. Takeuchi et. T{ Interprocess Communication in Concurrent PROLOG T} _ 008 H. Hirakawa T{ Chart Parsing in Concurrent PROLOG T} _ 013 H. Nishikawa et. T{ The Personal Sequential Inference Machine(PSI): Its Design Philosophy and machine architecture T} _ 016 M. Sato et. T{ Qute: A PROLOG/LISP Type Language for Logic Programming T} _ 017 T. Hikita T{ Average Size of Turner's Translation to Combinator Programs T} _ 018 H. Tamaki et. T{ A Transformation System for Logic Programs which Preserves Equivalence T} _ 019 H. Yasukawa T{ LFG in PROLOG --Toward a formal system for representing grammatical relations-- T} _ 020 H. Hirakawa et. T{ Implementing an OR-Parallel Optimizing PROLOG system (POPS) in Concurrent PROLOG T} _ 024 K. Sugiyama et. T{ A Knowledge Representation System in PROLOG T} _ 026 H. Yokota et. T{ An Enhanced Inference mechanism for Generating Relational Algebra Queries T} _ 027 H. Yasuura T{ On the parallel complexity of Unification T} _ 028 K. Sakai et. T{ Incorporating Naive Negation into PROLOG T} _ 029 K. Furukawa et. T{ Mandala: A Concurrent PROLOG Based Knowledge Programming Language/System T} _ 030 H. Enomoto et. T{ Paradigms of Knowledge Based Software System and Its Service Image T} _ 033 N. Ito et. T{ Parallel Inference Machine Based on the Data Flow Model T} _ 034 E. Shapiro T{ Systems programming in Concurrent PROLOG T} _ 035 N. Ito et. T{ Parallel PROLOG Machine Based on the Data Flow Model T} _ 041 M. Aso T{ Simulator of XP'S T} _ 042 R. Onai et. T{ An Approach to a Parallel Inference Machine Based on Control-Driven and Data-Driven Mechanisms T} _ 044 T. Chikayama et. T{ ESP Reference Manual T} _ 046 J. Tsuji et. T{ Dialogue Management in the Personal Sequential Inference Machine (PSI) T} _ .TE .TS box; cb s s s s c|c|c|c n|ltw(1i)|ltw(3i)|ltw(1.5). ICOT TECHNICAL REPORT _ NO. Author Topic Comment _ 053 S. Shibayama et. T{ A Relational Database Machine with Large Semiconducter Disk and Hardware Relational Algebra Processor T} _ 055 T. Hattori et. T{ SIMPOS: An Operating System for a Personal PROLOG Machine PSI T} _ 056 T. Hattori et. T{ The Concept and Facilities of SIMPOS Supervisor T} _ 057 S.Takagi et. T{ Overall Design of SIMPOS (Sequential Inference Machine Programming and Operating System) T} _ 058 F. Maruykama et. T{ PROLOG-Based Expert System for Logic Design T} _ 059 T. Hattori et. T{ The Concept and Facilities of SIMPOS File System T} _ 060 T. Yokomori T{ A Note on the Set Abstration in Logic Programming Language T} _ 061 T. Kurokawa et. T{ Coordinator -- the Kernel of the Programming System for the Personal Sequential Inference Machine (PSI) T} _ 062 K. Sakai T{ An Ordering for Term Rewriting Systems T} _ 063 H. Sakai et. T{ Design and Implementation of the Relational Database Engine T} _ 064 S. Shibayama et. T{ Query Processing Flow on RDBM Delta's Functionally-Distributed Architecture T} _ 065 K. Ueda et. T{ Efficient Stream/Array Processing in Logic Programming Language T} _ 066 K. Iwata et. T{ Design and Implementation of a Two-Way Merge-Sorter and its Application to Relational Database Processing T} _ 067 H. Enomoto et. T{ Natural Language Based Software Development System Tell T} _ 068 H. Enomoto et. T{ Formal Specification and Verification for Concurrent Systems by Tell T} _ 071 M. Sugimoto et. T{ Design concept for a Software Development Consultation System T} _ 072 T. Ida et. T{ Comparison of Closure Reduction and Combinatory Reduction Schemes T} _ 074 N. Miyazaki et. T{ An Overview of Relational Database Machine Delta T} _ 075 K. Taki et. T{ Hardware Design and Implementation of the Personal Sequential Inference Machine (PSI) T} _ 076 K. Furukawa et. T{ Mandala: A Logic Based Knowledge Programming System T} _ 084 K. Murakami et. T{ Architectures and Hardware Systems: Parallel Inference Machine and Knowledge Base Machine T} _ .TE .TS box; cb s s s s c|c|c|c n|ltw(1i)|ltw(3i)|ltw(1.5). ICOT TECHNICAL REPORT _ NO. Author Topic Comment _ 086 S. Uchida et. T{ Sequential Inference Machine: SIM Progress Report T} _ 088 H. Sawamura et. T{ Recursive Unsolvability of Determinacy, Solvable Cases of Determinacy and their Applications to PROLOG Opetimization T} _ 089 T. Kakuta et. T{ The Design and Implementation of Relational Database Machine Delta T} _ 090 T. Miyazaki et. T{ A Sequential Implementation of Concurrent PROLOG based on the Shallow Binding Scheme T} _ .TE ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Wed Jan 1 11:47:16 1986 Date: Fri, 20 Dec 85 02:19:48 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a022322; 18 Nov 85 13:22 EST Date: Mon 18 Nov 1985 09:24-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #172 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Mon, 18 Nov 85 22:08 EST AIList Digest Monday, 18 Nov 1985 Volume 3 : Issue 172 Today's Topics: Queries - Fictional Machines That Talk & Object-Oriented Programming, AI Tools - Typed Languages and Lisp, Psychology - Prejudice, Inference - Abduction & Translating Representations, Intelligence - IQ Test for AI, Review - TI's Satellite Symposium and AI Hype, News - RCA Chair at Penn ---------------------------------------------------------------------- Date: 15 Nov 85 16:17 PST From: Halvorsen.pa@Xerox.ARPA Subject: Fictional accounts of machines that talk I am looking for references to early (or ideally the earliest) mention of machines with natural language capabilities in fiction or film. ------------------------------ Date: Wed, 13 Nov 85 12:09 From: Nick Davies (at GEC Research) Subject: Object oriented programming in Common Lisp Does anyone have or know of an implementation of Flavors or any other object-oriented programming system in Common Lisp ? Nick Davies (ye85%uk.co.mrca@cs.ucl.ac.uk). Thanks in advance ------------------------------ Date: Sat, 16 Nov 85 14:09:10 gmt From: Don Sannella Subject: Re: Typed languages and Lisp >From: Skef Wholey A type checker can find some bugs, but it isn't clear that such bugs would take much time to find and fix relative the to the "real" bugs a programmer spends most of his time on. Also, actually entering type information can add to program development time. Controlled experiments are required ... I have been programming in HOPE and ML (both of which have the same strong but flexible type system, due to Robin Milner) for about seven years. I do not know of any controlled experiments which show that strong type checking decreases program development time; I can only report that in my experience using these languages, the type checker catches so many bugs (I would guess far in excess of 95% of non-syntax errors) that programs really do usually run correctly the first time they get past the type checker. I have heard other people who use these languages say the same thing, and I haven't met anybody who has really tried to use them complaining that entering type information doesn't pay off. (This is especially true of ML, where the compiler is able to infer types of functions so that a programmer is not required to say much about types at all.) I shudder to think of how much time I would have wasted in LISP trying to track down some of the (often subtle) type bugs the compiler has caught for me! Disclaimer: I haven't tried writing things like operating systems or compilers in a language like HOPE or ML (although other people have done so), so I don't know how much the strong type system gets in the way in cases like these. My experience is with rather mathematically-oriented programs of 1000 lines or so. ------------------------------ Date: Thu, 14 Nov 85 22:32:12 PST From: Richard K. Jennings Subject: Removing Prejudice Concerning removing prejudice, let me suggest reading "The Nature of Prejudice" by Gordon Allport and Patrick Monihan (then a Prof at Harvard). I think modeling prejudice accurately is the mandate for AI systems, not removing it. Rich. ------------------------------ Date: Fri, 15 Nov 85 03:16 EST From: jcma@MIT-MC.ARPA Subject: Abduction Makes The Big Time See Charniak and McDermott, Introduction to AI, Addison-Wesley, 1985 for several big sections on abduction. ------------------------------ Date: 15 Nov 85 10:27 PST From: Shrager.pa@Xerox.ARPA Subject: Abduction Isn't this what *all* script and schema inference systems do? Isn't this what *all* pattern recognition systems do? In fact almost every AI system that is not proof-based does something like what you are calling abduction. I can't understand why anyone thinks that this is news, other than having a fancy new name for what we've been doing for years. -- Jeff ------------------------------ Date: 17 Nov 85 21:18 PST From: Shrager.pa@Xerox.ARPA Subject: On translating representations "[...] for psychological reasons, in order to guess new theories, [mathematically equivalent theories] may be very far from equivalent, because one gives a man different ideas from the other. [...] every theoretical physicist who is any good knows six or seven different theoretical representations for exactly the same physics. He knows they are equivalent, and that nobody is ever going to be able to decide which one is right at that level, but he keeps them in his head, hoping that they will give him different ideas for guessing [new theories]." From: Feynman, R (1965). The Character of Physical Law. MIT press, Cambridge, Mass. pp. 168. ------------------------------ Date: Thu, 14 Nov 85 10:39:47 est From: Geoff Loker Subject: Re: IQ test for AI. (AIList Digest V3 #164) > I don't remember anyone suggesting to test AI programs with the > traditional human IQ test. [...] > Has any one tried to write such a computer program ? > Rene Bach (Bach@score) Thomas Evans' PhD thesis (MIT 1963) was concerned with the solution by machine of so-called "geometric-analogy" IQ test questions. The program he developped was called ANALOGY, and it apparently did quite well with the "A is to B as C is to __" picture problems. A revised form of his thesis can be found in "Semantic Information Processing", edited by Marvin Minsky, published 1968 by MIT Press. Also, the second edition of Winston's "Artificial Intelligence" (1984) refers to Evans' work and may have pointers to follow-up work as well. ------------------------------ Date: Tue, 12 Nov 85 12:04:21 EST From: Jakob Nielsen Subject: Re: Future IQ I have actually for some time used the term "AIQ" (Artificial Intelligence Quotient) as a shorthand when discussing computers (e.g. "this system has an incredibly low AIQ"). I don't think that it would have any meaning to use human IQ tests on computers however - just understanding the questions themselves (without trying to *answer* them) would require a higher AIQ than normally seen these days. Of course in 2085, who knows ... ------------------------------ Date: Fri, 15 Nov 1985 03:39 EST From: "David D. Story" Subject: TI's Satellite Symposium & AI Hype I went to this thing. It was terrible. The technical content was zilch. I also question the way that "expert systems" and rule-production systems are lumped together. Expert systems are defined as systems that can handle areas of knowledge in the manner that a human expert does. They may or may not be rule-production systems. Rule Production systems may not exhibit even vaguest qualities of a human expert. This type of lumpage will only serve to confuse the general public and lend the credence of the speakers and TI to real consumer fraud. There is not even mention of the performance criteria that originally generated the name of "EXPERT SYSTEM". This is further solidified by a apprx. 25 year old student saying that he wrote an "EXPERT SYSTEM" with a couple of rules. If I say that you have a gross income of over $6000 dollars and you have to file, does that make me an EXPERT. I think not. There is no mention of the degree of expertice of such systems should possess before being named even decision support tools. What compounds this was I attended the Press viewing of the symposium and there was no one there to field guestions. Most novices come with a pre-concieved idea of AI and leave being more reinforced with errant notions. This feeds into Geoff Goodfellow's previous posting on this list. The furtherence of this hype is just another blackeye AI researchers are going to have to bear when it comes to how much funding AI is receiving. Generating interest by speculation is one thing - opening the doors for real consumer fraud other ! Dave Story FTD@MIT-OZ ------------------------------ Date: Mon, 18 Nov 85 02:03 EST From: Tim Finin Subject: RCA Chair at Penn RCA ESTABLISHES AN AI CHAIR AT THE UNIVERSITY OF PENNSYLVANIA The University of Pennsylvania is pleased to announce the establishment of the RCA Professorship in Artificial Intelligence in the Department of Computer and Information Science. Established through the generosity of RCA, the chair is intended to enhance the University's position as a leading center for research and education in artificial intelligence. The University's tradition of pioneering work in the computing field, from the development of ENIAC, the offering of the first computer course, and the graduation of the first PhD in computer science, manifests itself today in a strong focus in the Computer and Information Science Department on artificial intelligence. The department is involved in a variety of research programs at the forefront of the field, including work in natural language processing, knowledge representation and reasoning, expert systems, computer vision and robotics, computer graphics, parallel processing, programming languages and environments, computer architecture and systems, theory of computation, software engineering, and database systems. These research programs involve interactions with the Departments of Electrical and Mechanical Engineering (in sensors and robotics), the Department of Chemical Engineering (in expert systems), the Departments of Linguistics, Philosophy, and Psychology (in cognitive science, in particular, in the computational aspects of language and perception), the Wharton School (in database systems, expert systems, and software engineering), and the Medical School (in expert systems and image processing). The person appointed to the RCA Professorship will be a distinguished scholar who has performed outstanding research in artificial intelligence with both theoretical and practical significance, especially in the areas of knowledge representation and reasoning with potential applications to the new generation of expert systems technology. ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Fri Dec 20 02:19:40 1985 Date: Fri, 20 Dec 85 02:19:16 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a028438; 19 Nov 85 3:57 EST Date: Tue 19 Nov 1985 00:15-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #173 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Tue, 19 Nov 85 06:32 EST AIList Digest Tuesday, 19 Nov 1985 Volume 3 : Issue 173 Today's Topics: Seminars - Adaptive Planning (UCB) & Sparse Distributed Memory (BBN) & Explanation-Based Learning (BBN) & Learning Search Control Knowledge (CMU) & MED2 Diagnostic Expert (MIT) & Truth Maintenance, Multiple Worlds in KEE (SU) & Representation of Natural Form (SU) & Setting Tables and Illustrations With Style (CSLI), Course - Connectionist Models (CMU) ---------------------------------------------------------------------- Date: Thu, 14 Nov 85 16:57:53 PST From: admin%cogsci@BERKELEY.EDU (Cognitive Science Program) Subject: Seminar - Adaptive Planning (UCB) BERKELEY COGNITIVE SCIENCE PROGRAM Fall 1985 Cognitive Science Seminar - IDS 237A Tuesday, November 19, 11:00 - 12:30 240 Bechtel Engineering Center Discussion: 12:30 - 1:30 in 200 Building T-4 ``Adaptive Planning is Commonsense Planning'' Richard Alterman Computer Science Division, U.C.B. A characteristic of commonsense planning is that it is knowledge intensive. For most mundane sorts of situations human planners have access to, and are capable of exploiting, large quantities of knowledge. Commonsense planners re-use old plans under their normal circumstances. Moreover, commonsense planners are capable of refitting old plans to novel cir- cumstances. A commonsense planner can plan about a wide range of phenomena, not so much because his/her depth of knowledge is consistent throughout that range, but because s/he can re-fit old plans to novel contexts. This talk is about an approach to commonsense planning called adaptive planning. An adaptive planner plans by exploit- ing planning knowledge in a manner that delays the reduction of commonsense planning to problem-solving. Key elements in the theory of adaptive planning are its treatment of background knowledge and the introduction of a notion of planning by situation matching. This talk will describe adaptive planning as it applies to a number of commonsense planning situations, including: riding the NYC subway, trading books, transferring planes at JFK airport, and driving a rented car. ------------------------------ Date: 14 Nov 1985 11:48-EST From: BGOODMAN at BBNG.ARPA Subject: Seminar - Sparse Distributed Memory (BBN) [Forwarded from the MIT bboard by SASW@MIT-MC.] BBN Labs SDP AI Seminar Speaker: Dr. Michael R. Raugh Research Institute for Advanced Computer Science NASA Ames Research Center Title: Kanerva's Sparse Distributed Memory: A RIACS Project Date: Friday, November 22nd, 2:00pm Location: Main Seminar Room (2nd floor) Bolt Beranek and Newman Inc. 50 Moulton Street Cambridge, MA. An exciting new concept in which information is stored in a large number of neighboring addresses determined by "content," produces a memory that retrieves causal relationships as well as sequences of episodes and is sensitive to similarity. It is also forgetful and reinforcable: a memory much like yours and mine. ------------------------------ Date: 14 Nov 1985 11:48-EST From: BGOODMAN at BBNG.ARPA Subject: Seminar - Explanation-Based Learning (BBN) [Forwarded from the MIT bboard by SASW@MIT-MC.] BBN Labs SDP AI Seminar Speaker: Professor Gerald DeJong Coordinated Science Laboratory University of Illinois at Urbana-Champaign Title: Explanation Based Learning Date: Monday, November 25th, 10:30am Location: 2nd Floor Large Conference room BBN Laboratories Inc. 10 Moulton Street Cambridge, MA. Machine learning is one of the most important current areas of artificial intelligence. With the trend away from "weak methods" and toward a more knowledge intensive approach to intelligence, the lack of knowledge in an AI system becomes one of the most serious limitations. This talk advances a technique called explanation based learning. It is a method of learning from observation. Basically, it involves endowing a system with sufficient knowledge so that intelligent planning behavior of others can be recognized. Once recognized, these observed plans are generalized as far as possible while preserving the underlying explanation of their success. The approach supports one-trial learning. A new general concept can be acquired from an observation of just one observed example. The approach has been applied to three diverse areas: natural language processing, robot task planning, and proof of propositional calculus theorems. The approach holds promise for solving the knowledge collection bottleneck in the construction of current knowledge-based systems. ------------------------------ Date: 14 Nov 85 23:39:59 EST From: Steven.Minton@CAD.CS.CMU.EDU Subject: Seminar - Learning Search Control Knowledge (CMU) [Forwarded from the CMU bboard by Laws@SRI-AI.] On Wednesday, November 20, at 12:00 I will present my thesis proposal in 5409. My thesis is concerned with the use of explanation-based generalization in the PRODIGY system, a learning apprentice that (among other things) acquires search control rules. The title is: "Analytic Techniques for Learning Search Control Knowledge". Copies are in the lounge. ABSTRACT Compression analysis, the subject of the proposed thesis, is a method for analyzing search spaces to produce effective search control rules. As with previous explanation-based learning techniques, an example problem focuses the analysis process so that the entire search space need not be analyzed. The key idea behind compression analysis is that many alternative explanations can be produced to justify a search control decision; therefore it is appropriate to search for an explanation that produces the most effective generalized control rule. In practice this is achieved by proposing an initial explanation which is then improved using a set of heuristic transformation strategies. ------------------------------ Date: Sun, 17 Nov 85 16:28:06 EST From: "Steven A. Swernofsky" Subject: Seminar - MED2 Diagnostic Expert (MIT) Wednesday 20, November 4: 00pm (4:15 Refreshments) Room: NE43-512A "MED2: An Expert System Shell for Diagnosis and Therapy in Complex Domains" Frank Puppe Kaiserlautern University Germany Concentrating on the medical domain, MED2 is a shell combining a wide variety of important aspects of clinical reasoning. It's "Working-Memory" control structure involves investigating a set of hypotheses simultaneously, avoiding the shortcomings of focussing on the top-hypothesis only. This concept allows using differential diagnosis techniques and exploiting relationships among patho-concepts in an efficient manner. Other interesting features of MED2 include separation of database and diagnostic reasoning, temporal reasoning, and belief revision. HOST: Prof. Peter Szolovits ------------------------------ Date: Mon 18 Nov 85 08:32:01-PST From: Anne Richardson Subject: Seminar - Truth Maintenance, Multiple Worlds in KEE (SU) DAY December 3, 1985 EVENT Computer Science Colloquium PLACE Skilling Auditorium TIME 4:15 TITLE "Truth Maintenance and Multiple Worlds in KEE" PERSON Paul Morris, Robert Nado, Richard Fikes FROM IntelliCorp TRUTH, MAINTENANCE AND MULTIPLE WORLDS IN KEE We describe the integration of an assumption-based truth maintenance system (ATMS) into the frame-based representation facilities of the KEE system, and the use of the ATMS to implement a multiple-world context graph system for KEE. Integration into the frame system involves associating with potential slot values ATMS nodes that are used to determine in which worlds (contexts) the slot values are believed. Built-in inferences provided by the frame system, such as inheritance and the checking of value class and cardinality constraints, are recorded, when needed, as explicit justifications in the ATMS. In addition, the default reasoning capabilities of KEE have been refined and extended to take advantage of the ATMS. Tradeoffs in the integration between flexibility of use and run-time efficiency are examined. We describe the multiple-world context graph system with particular attention to an interpretation of the graph as a network of actions. In this framework, the semantics of graph merges are investigated and restrictions to ensure valid action sequences are discussed. ------------------------------ Date: Mon 18 Nov 85 08:29:15-PST From: Anne Richardson Subject: Seminar - Representation of Natural Form (SU) DAY November 19, 1985 EVENT Computer Science Colloquium PLACE Skilling Auditorium TIME 4:15 TITLE Perceptual Organization and the Representation of Natural Form PERSON Alex P. Pentland FROM AI Center, SRI Int'l and CSLI, Stanford PERCEPTUAL ORGANIZATION AND THE REPRESENTATION OF NATURAL FORM To understand both perception and commonsense reasoning we need a representation that captures important physical regularities and that correctly describes the people's perceptual organization of the stimulus. Unfortunately, the current representations were originally developed for other purposes (e.g., physics, engineering) and are therefore often unsuitable. We have developed a new representation and used it to make accurate descriptions of an extensive variety of natural forms including people, mountains, clouds and trees. The descriptions are amazingly compact. The approach of this representation is to describe scene structure in a manner similar to people's notion of ``a part,'' using descriptions that reflect a possible formative history of the object, e.g., how the object might have been constructed from lumps of clay. For this representation to be useful it must be possible to recover such descriptions from image data; we show that the primitive elements of such descriptions may be recovered in an overconstrained and therefore reliable manner. An interactive ``real-time'' 3-D graphics modeling system based on this representation will be shown, together with short animated sequences demonstrating the descriptive power of the representation. ------------------------------ Date: Mon 18 Nov 85 11:46:58-PST From: Fred Lakin Subject: Seminar - Setting Tables and Illustrations With Style (CSLI) Pixels and Predicates: SETTING TABLES AND ILLUSTRATIONS WITH STYLE Who: Rick Beach, Xerox PARC Where: CSLI trailers When: 1:00pm - Wednesday, November 20, 1985 Abstract: Two difficult examples of incorporating complex information within electronic documents are illustrations and tables. The notion of style, a way of maintaining consistency, helps manage the complexities of formatting both tables and illustrations. The concept of graphical style extends document style to illustrations. Observing that graphical style does not adequately deal with the layout of information leads to the study of formatting tabular material. A grid system for describing the arrangement of information in a table, and a constraint solver for determining the layout of the table are key components of this research. These ideas appear to extend to formatting other complex material, including mathematical typesetting and page layout. Several typographic issues for illustrations and tables will be highlighted during the talk. ------------------------------ Date: 18 Nov 85 23:29 EST From: Dave.Touretzky@A.CS.CMU.EDU Subject: Course - Connectionist Models (CMU) CONNECTIONIST MODELS: A SUMMER SCHOOL Sponsored by the Sloan Foundation ORGANIZERS: Geoffrey Hinton (Carnegie-Mellon University) Terrence Sejnowski (The Johns Hopkins University) David Touretzky (Carnegie-Mellon University) DATE: June 20 through 29, 1986 PLACE: Carnegie-Mellon University, Pittsburgh, Pennsylvania PURPOSE OF THE PROGRAM: The purpose of the summer school is to familiarize young researchers with current techniques in the area of connectionist models of intelligence. This includes search procedures, learning procedures, and methods for representing knowledge in massively parallel networks of neuron-like units. Application areas include vision, speech, associative memory, natural language and motor control. FACULTY: There will be six full time Tutors plus several Guest Lecturers. Tutors: Guest Lecturers: James Anderson, Brown University Jerome Feldman, U. of Rochester Dana Ballard, U. of Rochester Christof Koch, MIT Andrew Barto, U. Mass. Amherst David Rumelhart, UCSD Geoffrey Hinton, CMU David Touretzky, CMU James McClelland, CMU others to be announced Terrence Sejnowski, Johns Hopkins WHO MAY ATTEND: Participation is limited to graduate students and recent PhD's who are or will be working on connectionist models. About 40 students will be accepted. Persons who have already completed a Ph.D. degree must have done so no earlier than January 1985 to be eligible to attend. EXPENSES: There is no tuition charge. Funding from the Sloan Foundation will provide dormitory accommodations and breakfast and lunch for each attendee, plus reimbursement for a substantial portion of travel expenses. HOW TO APPLY: By March 1, 1986, send your curriculum vitae and a copy of one relevant paper, technical report, or research proposal to: Dr. David Touretzky, Computer Science Department, Carnegie-Mellon University, Pittsburgh, PA, 15213. Applicants will be notified of acceptance by April 15, 1986. ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Wed Jan 1 12:31:16 1986 Date: Wed, 1 Jan 86 12:31:04 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a013463; 20 Nov 85 13:53 EST Date: Wed 20 Nov 1985 09:53-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #174 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Sat, 23 Nov 85 01:48 EST AIList Digest Wednesday, 20 Nov 1985 Volume 3 : Issue 174 Today's Topics: Queries - PEARL & Man/Machine IQ Tests & Smalltalk, AI Tools - Object-Oriented Programming, Science Fiction - Machines Who Talk, Survey - Classic AI Books, AI Tools - Xerox 1186 ---------------------------------------------------------------------- Date: 20 NOV 85 12:09-N From: SCHNEIDER%CGEUGE51.BITNET@WISCVM.ARPA Subject: Information about PEARL Does anybody know, whether the PEARL project at Berkeley is still alive and whom I could contact if I have questions ? (I couldn't reach anybody of the persons mentioned in the Pearl manual distributed with the UNIX Franz Lisp.) I also would be glad to get some comments on that language or pointers to systems that have been built with it. SNAILMAIL: Daniel Schneider Departement de science politique Universite de Geneve 1211 GENEVE 4 (=GENEVA) Switzerland BITNET: SCHNEIDER@CGEUGE51 >From ARPANET: SCHNEIDER%CGEUGE51.BITNET@WISCVM Usenet: mcvax!cernvax!cui!shneider (shneider without "c") ------------------------------ Date: Mon, 18 Nov 85 14:38:58 PST From: Laursen.pa@Xerox.ARPA Subject: Re: IQ test for AI. (AIList Digest V3 #164) Coming at this from another angle, has anyone considered building an AI system that boosts a human being's performance on IQ tests? Can a person/computer combination confirm that 'Two heads are better than one?" A baseline for comparison would have to allow the non-AI assisted person access to a calculator and an on-line dictionary while taking an IQ test. How much leverage beyond that could an 'intelligent' system give? ------------------------------ Date: nov 20 1985 From: astropa%ipacuc.bitnet@WISCVM.ARPA Subject: smalltalk I would like to know if anyone can help me to reach the people who originally implemented Smalltalk-80 at Xerox on the networks. They are Adele Goldberg, David Robson, Daniel Ingalls. I also would like to know if there is somebody out there interested in Smalltalk, and if someone knows about public-domain implementations of Smalltalk. Please send private replies to me. Fabio Favata Net address: ASTROPA@IPACUC.BITNET Use the Wisconsin ARPA-BITNET gateway to reach Bitnet. ------------------------------ Date: 18 Nov 85 15:44 PST From: Bobrow.pa@Xerox.ARPA Subject: Object-Oriented Programming in Lisp From: Nick Davies (at GEC Research) Subject: Object oriented programming in Common Lisp > Does anyone have or know of an implementation of Flavors > or any other object-oriented programming system in Common Lisp ? CommonLoops is an object oriented language that is written in Common Lisp. It is being proposed by Xerox as a standard for object-oriented programming for that language. CommonLoops merges the facilities of object oriented programming and Lisp. It has a unique combination of properties. These include a uniform syntax for function calling and sending messages; a merger of the type space of Lisp and the class hierarchy of objects; a generalization of method specification that includes ordinary Lisp functions at one extreme, and fully type specified functions at the other; and a "metaclass" mechanism that allows tradeoffs between early binding and ease of exploratory programming in the implementation of objects. For information about the availability of an experimental version of CommonLoops in Common Lisp contact Paul Ricci (Ricci@Xerox.ARPA). For a copy of a paper describing CommonLoops, please send your U.S. Mail address to MGardner@Xerox. ------------------------------ Date: Mon, 18 Nov 85 20:55:28 EST From: "Keith F. Lynch" Subject: Re: Fictional accounts of machines that talk Date: 15 Nov 85 16:17 PST From: Halvorsen.pa@Xerox.ARPA I am looking for references to early (or ideally the earliest) mention of machines with natural language capabilities in fiction or film. In the story of Theseus and the Minotaur, author and age (at least 3000 years) unknown, Talus, the man of brass, is a minor character. It was hammered out by Vulcan, owned by King Minos, and drowned when it overreached itself attempting to club Theseus' vessel. It was able to speak Greek. ...Keith ------------------------------ Date: Mon, 18 Nov 85 17:43:08 est From: Walter Hamscher Subject: Machines Who Talk Try Pam McCorduck's "Machines Who Think". The first chapter "Brass for Brains" tells the history of AI from ancient Greece to the Dartmouth conference. Page 4 has a quote from Homer's Iliad, circa 850 BC, concerning Hephaestus' attendants, whom he created in his forge: These are golden, and in appearance like living young women. There is intelligence in their hearts, and there is speech in them and strength, and from the immortal gods they have learned how to do things. ------------------------------ Date: Mon, 18 Nov 85 14:47 EST From: Seth Steinberg Subject: Old Time Natural Language@bbn-vax.ARPA How about Roger Bacon's great bronze head? He supposedly built it to answer questions for him which it supposedly did rather well. It finally said, "Time is, time was, time past." and shattered into pieces but we've all had this happen to us at one time or another. Before you get too impressed with this story think of how much more impressed you'd be if Bacon had supposedly typed in the questions on a keyboard? There might be older stories lurking around Daedalus or possibly in Ezekiel which supposedly has something for everyone in it. Seth Steinberg ------------------------------ Date: Tuesday, 19 November 1985 17:50:28 EST From: Dan.Miller@a.sei.cmu.edu Subject: Results - Classic AI Books Survey The following are the results of an October 19, 1985 AIList survey concerning classic AI books. Thanks to all who responded. --- Daniel "Dan" H. MIller Software Engineering Institute dhm@sei.cmu.edu (dhm@cmu-sei.arpa) Carnegie-Mellon University (412)578-7700 Pittsburgh, PA 15213 USA "Disclaimer: The views and conclusions are those of the author, and should not be interpreted as representing the official policies, either expressed or implied, of any organization he may be affiliated with." =========================================================================== CLASSIC ARTIFICIAL INTELLIGENCE BOOKS SURVEY RESULTS =========================================================================== Total people responding: 5 Total number of responses: 14 ........................................................................... Number of responses for the same book: 2 Minsky, "Semantic Information Processing" MICHAELG%MIT-OZ@MIT-MC.ARPA GLICK@AIDS-UNIX.ARPA ........................................................................... Multiple responses for same author: 2 sets Nilsson, "Principles of AI". TIM%UPENN.CSNET@CSNET-RELAY Nilsson, "Problem Solving Methods in AI" GLICK@AIDS-UNIX.ARPA Winston, "The Psychology of Computer Vision" MICHAELG%MIT-OZ@MIT-MC.ARPA Winston, "Artificial Intelligence" MICHAELG%MIT-OZ@MIT-MC.ARPA (the best intro to the field) ........................................................................... Number of responses for single books: 8 Bobrow & Collins, eds., "Representation and Understanding". Academic Press, 1975. Still in print. VIS!GREG@SDCSVAX.ARPA Duda & Hart, "Pattern Classification & Scene Analysis" (also a classic in the pattern recognition literature) MICHAELG%MIT-OZ@MIT-MC.ARPA Feigenbaum and Feldman, "Computers and Thought" GLICK@AIDS-UNIX.ARPA Hofstader (sp?), Douglas, "Godel, Escher, Bach: An Eternal Golden Braid" (good reading for people who are AI oriented) PES@MITRE-BEDFORD.ARPA Marr, "Vision" MICHAELG%MIT-OZ@MIT-MC.ARPA (published in 82, this book may be too young to be considered a classic, but it expounds a set of principles that will always be applicable to vision) Minsky & Papert, "Perceptrons" MICHAELG%MIT-OZ@MIT-MC.ARPA (also a classic in the pattern recognition literature) Newell & Simon, "Human Problem Solving" MICHAELG%MIT-OZ@MIT-MC.ARPA (a thorough analysis containing many important ideas) Weizenbaum, "Human Reason: From Judgement to Calculation" (a very readable book with some very powerful ideas) PES@MITRE-BEDFORD.ARPA ------------------------------ Date: Mon 18 Nov 85 18:55:21-PST From: Christopher Schmidt Subject: Re: Comments on the Xerox 1186... We've had a Xerox 1186 here for a couple of weeks running a beta-version of Koto (the next release of Interlisp-D). It is configured with 3.8 Mb of memory, a 40 Mb disk, and the smaller display (633x832). The sales lit notes that you can buy as little memory as 1.1 Mb, disks sized at 10 Mb, 20 Mb, 40 Mb or 80 Mb, and a 15 or 19 inch display.) I've used as little as 1.1 Mb of memory in a dolphin and found that a typical Interlisp-D working set wants more. I recommend at least 1.6 Mb. Similarly, a 10 Mb disk isn't large enough to support much virtual memory. I recommend at least 20 Mb. We've found the small display adequate, but, naturally, prefer more bits. (The 19 inch screen has more bits on it than the 808x1024 bits on the 17 inch display found on dolphins and dandelions, but I don't know how many.) Before laying eyes on an 1186 we ordered a number of them for program development with 3.8 Mb of memory, 40 Mb of disk, and the 19 inch display. From our recent experience I wouldn't consider changing the configuration in any respect. The processor is less than half the size of a dandelion and a good deal cooler and quieter, but still not 100% unobtrusive. The keyboard is a major improvement over that of the dandelion. In speed it is not markedly different from a dandetiger, doing some things a little faster and others a little slower. Eg. it reads files off of our unix file server slightly faster than a 'tiger and off a Xerox 8037 slightly slower. The discrepancies that favor one machine over the other are roughly equal in number and not more than about 10-20% in magnitude. Exceptions: It's much weaker in floating point (it lacks the special hardware), and in drawline (it lacks the microcode). I expect the microcode situation to improve, but I have been unable to get a statement out of AIS on this point. I'd be willing to answer specific questions, but, as noted above, we've only had it for a couple of weeks and we're running beta-release software, so "your mileage may vary." --Christopher ------------------------------ Date: 19 Nov 85 09:42 PST From: Stern.pasa@Xerox.ARPA Subject: Re: Comments on the Xerox 1186... Just a note from the XAIS sales staff: the 1.1 mb RAM and 10 mb disk are offered for "delivery systems" use, where an end user is not doing development, and is probably running only compiled code. The average application, even in runtime-only mode, may well require more of each resource, but then again it may not. For development, you probably want the full-up machines. Josh ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Wed Jan 1 12:30:57 1986 Date: Wed, 1 Jan 86 12:30:45 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a029650; 22 Nov 85 1:42 EST Date: Thu 21 Nov 1985 21:52-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #175 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Sat, 23 Nov 85 02:09 EST AIList Digest Friday, 22 Nov 1985 Volume 3 : Issue 175 Today's Topics: Seminars - Unification Revisited (SRI) & Expanding the Horizons of Expert Systems (SRI) & CYC Commonsense Knowledge Project (GTE) & A Multimodal Perceptual System (UPenn) & ANALOGICA '85 (Rutgers), Conference - Foundations of AI ---------------------------------------------------------------------- Date: Tue 19 Nov 85 10:02:51-PST From: OLENDER@SRI-AI.ARPA Subject: Seminar - Unification Revisited (SRI) DATE: Monday, November 25, 1985 LOCATION: EJ242 UNIFICATION REVISITED Jean-Louis Lassez IBM Thomas J. Watson Research Center There are three main approaches to finitely represent sets of solutions of equations in the Herbrand Universe. In Robinson's classical approach the set of solutions is represented by an mgu which is computed from the set of equations. We introduce a dual approach, based on Plotkin's and Reynold's concept of anti-unification in which the finite representation (mgs) is now "lifted" from the set of solutions. A third approach proposed by Colmerauer is based on the concept of eliminable variables. The relationships between these three approaches are established. This study provides an appropriate setting to address the problem of solving systems of equations and inequations which arises in recent extensions to Prolog. A key result is that the meta-equation E = E1 v E2 v ... v En admits solutions only in trivial cases. Two important corollaries follow naturally. The first is Colmerauer's property of independences of inequations. This means that deciding whether a system of equations and inequations has solutions can be done in parallel. The other corollary is a negative result; the set of solutions of a system of equations and inequations can be finitely represented by mgu's only in trivial cases. Consequently, one cannot obtain a simplified system which is in "solved" form. This is unlike the case when only equations are considered. Similar properties hold in inductive inference when one attempts to generalize from sets of examples and counter-examples. ------------------------------ Date: Tue 19 Nov 85 15:25:48-PST From: ICHIKI@SRI-AI.ARPA Subject: Seminar - Expanding the Horizons of Expert Systems (SRI) EXPANDING THE HORIZONS OF EXPERT SYSTEMS Piero P. Bonissone and Allen L. Brown, Jr. General Electric Corp. R&D Conference Room EK242 Thursday, November 21, 1985 (10:00 - 11:00a.m.) Abstract: In this paper we analyze the complexity of problem domains, such as maintenance problems, typically handled by first generation expert systems. DELTA/CATS-1, an expert system for troubleshooting diesel electric locomotives, is described as a typical example of such systems. More complex domains of expertise involving time-varying, partial and uncertain information cannot be addressed by the techniques common to first generation expert systems. LOTTA, a symbolic simulator of battlefield situations, illustrates some of the requirements of such an intricate domain. We discuss the new inference techniques required to address the problem of managing battlefield strategies and tactics. These techniques include capabilities for reasoning with uncertain and incomplete information. The state of the art and current research thrusts are discussed. ------------------------------ Date: Thu, 21 Nov 85 11:16:19 EST From: Bernard Silver Subject: Seminar - CYC Commonsense Knowledge Project (GTE) GTE Laboratories AI Seminar Monday, November 25, 3 pm Room 8-2335 GTE Laboratories 40 Sylvan Rd.,Waltham, MA Mayank Prakash of MCC, Austin, Texas, will discuss the current state of research on the CYC Project. The CYC Project The CYC Project at MCC is an attempt to enhance the power of AI systems by providing them with common sense knowledge. Our approach is based on the observations that 1) the brittleness of current AI systems is due to a lack of knowledge of the broader context of their narrow domain, and 2) automated acquisition of knowledge requires the system to start out with a critical mass of general knowledge. Our goal is to create a system with a large common sense knowledge base. It will serve as a source of power for other AI systems by acting as a knowedge server and thus providing a deeper understanding of the domain, and by providing the ability to analogise. It is our belief that this will significantly enhance the performance of such systems. For more information, please contact Bernard Silver (617) 466-2663 ------------------------------ Date: Thu, 21 Nov 85 12:11 EST From: Tim Finin Subject: Seminar - A Multimodal Perceptual System (UPenn) Forwarded From: Sharon Stansfield Date: Thu, 21 Nov 85 11:09 EST THESIS PROPOSAL: A RUDIMENTARY ACTIVE, MULTIMODAL, INTELLIGENT SYSTEM FOR OBJECT CATEGORIZATION S. A. Stansfield This proposal outlines the design of a knowledge-based, active, multimodal perceptual system for the task of object categorization. Passive stereo vision and active touch are used. Basic level, or generic, objects will be recognized. The representation for reasoning will be hierarcical and frame-based. At the highest level will reside a representation of the object as a whole, including the features which comprise it and the relations among them. Intermediate levels will contain frames for the various features and the slots to be filled will contain the tactile, visual, and amodal properties of these features. The geometric representation will be based upon the idea of a spatial polyhedron: an object centered guide to the exploration of the object. The architecture will be a distributed hierarchy of knowledge-based modules, each domain specific and informationally encapsulated. These experts will be dedicated to the exploration for and identification of the features recognized by the system. Each module will be responsible for filling in the slots of the frame with which it is associated. At the lowest level of the hierarchy are the visual and tactual perception systems. At the highest level are the two strategists: the reasoner, responsible for hypotheses generation, disambiguation, and culling; and the explorer, responsible for generating sensing strategies once a goal has been formulated by the reasoner. Advisor: Committee: R. Bajcsy S. Lederman R. Paul L. Shastri Monday, 25 November 1985 10:00 a.m. Room To Be Announced ------------------------------ Date: 20 Nov 85 12:21:19 EST From: PRIEDITIS@RED.RUTGERS.EDU Subject: Seminars - ANALOGICA '85 (Rutgers) You are cordially invited to attend ANALOGICA '85 AT RUTGERS Monday, December 2, 1985 8:30 AM - 6:00 PM Hill Center for the Mathematical Sciences, Room 705 Busch Campus, Rutgers University New Brunswick, NJ Sponsored by Department of Computer Science Rutgers University, New Brunswick, NJ and GTE Fundamental Research Laboratories Waltham, MA Free and Open to the Public Analogica '85 is a multidisciplinary seminar on analogical reasoning, bringing together researchers from various disciplines such as artificial intelligence, cognitive psychology, linguistics, and philosophy of science. This one-day seminar will include eight half-hour talks and a panel discussion. Complimentary refreshments and lunch will be served. For more information, contact: Armand Prieditis, prieditis@rutgers.arpa, 201-932-4273 Smadar Kedar-Cabelli, kedar-cabelli@rutgers.arpa 201-932-4648 Tom Mitchell, faculty sponsor, 201-932-4716 SCHEDULE: 8:30 - 9:00 Coffee and donuts 9:00 - 9:15 Welcome 9:15 - 9:45 Dedre Gentner, University of Illinois at Urbana-Champaign 9:45 - 10:15 Mark Burstein, BBN Laboratories 10:15 - 10:30 Coffee and donuts 10:30 - 11:00 Paul Thagard, Princeton University 11:00 - 11:30 Russell Greiner, University of Toronto 11:30 - 12:00 Smadar Kedar-Cabelli, Rutgers University 12:00 - 2:30 Luncheon 2:30 - 3:00 Lindley Darden, University of Maryland 3:00 - 3:30 Bipin Indurkhya, Boston University 3:30 - 4:00 Keith Holyoak, University of Michigan 4:00 - 4:15 Coffee and Donuts 4:15 - 5:45 Panel discussion ------------------------------ Date: Mon, 18 Nov 85 17:40:07 mst From: yorick%nmsu.csnet@CSNET-RELAY.ARPA Subject: Conference - Foundations of AI WORKSHOP ON THE FOUNDATIONS OF AI Holiday Inn, Las Cruces, New Mexico Thursday February 6th - Saturday February 8th, 1986 Wednesday, February 5th: 6.00 pm welcome reception 9.00 - 9.30am: WELCOME AND INTRODUCTION: WILKS 9.30 - 12.30: LOGICAL/PHILOSOPHICAL FOUNDATIONS OF AI (1) (Chair: Wilks) CHANDRASEKARAN - Paradigms in AI: An historical and contemporary perspective HALPERN* - Turing's Test and the ideology of AI Coffee (15 mins) NEEDHAM - There's nothing special about AI General Discussion (30 mins) 2.00 - 5.00pm: LOGICAL/PHILOSOPHICAL FOUNDATIONS OF AI (2) (Chair: Hayes, Pat) DREYFUS - Traditional AI: A degenerating research program FODOR - Why there still has to be a language of thought Coffee (15 mins) DENNETT - The myth of original intentionality General Discussion (30 mins) Friday, February 7th: 9.30 - 12.30: RELATION BETWEEN FOUNDATIONS AND PROGRAMS (1) (Chair: Sleeman) BUNDY - What Kind of Field is AI? McCARTHY - AI Reasoning Should be Logic with Extensions Coffee (15 mins) CAMPBELL* - Novelties of AI: Theories, Programs, and Rational Reconstructions General Discussion (30 mins) 2.00 - 5.00pm: RELATION BETWEEN FOUNDATIONS AND PROGRAMS (2) (Chair: Brachman) SPARCK-JONES - What is an experiment in AI? NEWELL - On Comparing General Cognitive Architectures Coffee (15 mins) PYLYSHYN - Programs as models: Can there be a strong equivalence? General Discussion (30 mins) 6.00 pm An evening in Mexico Saturday, February 8th: 9.30 - 12.30: AI AND OTHER DISCIPLINES (1) (Chair: Uhr) ARBIB - How does Brain Theory Relate to AI? CHURCHLAND - AI and the neurosciences Coffee (15 mins) JOSHI - AI and Linguistics General Discussion (30 mins) 2.00 - 5.00: AI AND OTHER DISCIPLINES (2) (Chair: Schvaneveldt) RUMELHART - AI: What can Psychology learn? BODEN - Has AI helped Psychology? Coffee (15 mins) MINSKY - AI and Cognitive Science General Discussion (30 mins) 7.30pm: WORKSHOP BANQUET (Chair: Ortony) Sunday, February 9th: Depart for home, skiing etc. Each presenter will have 30 minutes to present his/her position followed by 15 minutes allocated for discussion. * Indicates submitted paper. Contact: derek@nmsu.csnet or yorick@nmsu.csnet ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Fri Dec 20 01:58:58 1985 Date: Fri, 20 Dec 85 01:58:36 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a005292; 22 Nov 85 15:09 EST Date: Fri 22 Nov 1985 10:51-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #176 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Sat, 23 Nov 85 02:23 EST AIList Digest Friday, 22 Nov 1985 Volume 3 : Issue 176 Today's Topics: Queries - Ordering ICOT Reports & MTSLISP on VM/CMS & Expert System Shells for IBM PC & Knowledge Ownership, Intelligence - Man/Machine IQ Tests, Programming Languages - Hard Typing, Literature - Spang Robinson Report, Expert Systems - COBOL Restructuring ---------------------------------------------------------------------- Date: Fri, 22 Nov 85 12:52:35 EST From: Frank Ritter Subject: Ordering ICOT Reports How does one order ICOT tech reports? [I'm haven't had time to really go though the list of reports that I had copied, but I'm sure many will want the address to order reports.] Thanks for posting the list, Frank ------------------------------ Date: Wed, 20 Nov 1985 18:42 PLT From: George Cross Subject: MTSLISP on VM/CMS Hi, Does anyone have MTSLISP running on VM/CMS? ---- George George R. Cross cross@wsu.CSNET Computer Science Department cross%wsu@csnet-relay.ARPA Washington State University faccross@wsuvm1.BITNET Pullman, WA 99164-1210 (509)-335-6319/6636 ------------------------------ Date: 20 Nov 85 15:03 PST From: kwhite.pasa@Xerox.ARPA Reply-to: kwhite.pasa@Xerox.ARPA Subject: Expert System Shells for IBM PC Does anyone out there have some experience on the IBM PC (or its compatible clones) using Expert System Shells? I have been giving consideration to KDS, EXSYS, and the Personal Consultant. I would like to have the system be capable of running DOS commands which could invoke other programs (in another language) and return. I am also particularly interested in shells that support the user when s/he is lost and says "I don't know" or "I don't understand". The system should be capable of making suggestions to the user in this situation. Thanks in advance for your help. ...Kendall... ------------------------------ Date: Wed, 20 Nov 85 12:02:13 est From: mayerk%UPenn-GradEd%upenn.csnet@CSNET-RELAY.ARPA Subject: Issues Concerning Expert Systems -- Who Owns What This article was clipped from the Wall Street Journal, Thursday, November 14, 1985. * * * AFTER 15 YEARS of medical training in Pittsburgh, Caduceus will soon be ready for its first patients. Caduceus is a computer program that tries to copy the skill of 72-year-old internist Jack Myers and other physicians at diagnosing about 600 illnesses. The computer will never be as good as the best doctor because it lacks an imagination, Dr. Myers says. Diseases that haven't been described or categorized will stump Caduceus, he explains, whereas "a brilliant human being will realize he's facing something new." Unlike a human being, however, Caduceus never forgets to order a test or overlook a symptom, so doctors could use the program to check their diagonoses. University of Pittsburgh computer engineer Harry Pople says the long time spent developing Caduceus is partly a result of the system's complexity: The software had to be designed to rapidly diagnose several illnesses at once. But the system had some human problems, too. An early version, call Internist I, was rather authoritarian, making diagnoses without telling doctors how it had arrived at its conclusions. That sort of bedside manner invariably stepped on fragile medical egos. Caduceus, named for the coiled-snake medical emblem, now reasons along with a doctor, suggests alternative diagnoses and explains its "reasoning." The system will be field-tested late next year. * * * I now would like to ask for opinions on the following issue; One reason experts (human) are valuable is because they are scarce. This scarcity is one of the driving forces behind implementing an expert system. My question then, Is knowledge proprietary, if so, who owns it? The domain expert, or the organization (university, corporation, etc.), that the domain expert worked for when the special knowledge was acquired? Is the knowledge "stored" within a domain experts' brain copyrightable? Obviously an expert system is, the same as any piece of software. But since, as so many people have said, "In the knowledge lies the power," an expert system is only as valuable as the domain knowledge it contains. I think that in the next few years, serious questions like these will affect the way we think about computer systems. Another hard question is who is liable? Let's say some expert system in medical diagnosis was fed data, which is later found to be faulty, or downright wrong. And a serious injury was related to a physician's use of this information. If the writers of the expert system could be shown to be negligent in the verification of their data, could it be possible that they are liable, just as if it were a case of malpractice? Kenneth Mayer (mayerk@UPenn-Graded) ------------------------------ Date: 21 Nov 85 07:35:42 EST (Thursday) From: MJackson.Wbst@Xerox.ARPA Subject: Re: IQ test for AI >From Laursen.PA@Xerox in V3 #174: ". . .has anyone considered building an AI system that boosts a human being's performance on IQ tests?" Of course, *any* computer system which is more than a toy is intended to boost human performance to some degree (answer questions one could not answer so quickly, or at all, unaided). Closer to the point, the unifying principle of Doug Engelbart's work over the years has been "augmentation of the human intellect." Mark ------------------------------ Date: Thu, 21 Nov 85 9:13:49 PST From: Laursen.pa@Xerox.ARPA Subject: Re: IQ test for AI "MJackson.Wbst's message of Thu, 21 Nov 85 7:35:42 EST" "...Of course, *any* computer system which is more than a toy is intended to boost human performance to some degree (answer questions one could not answer so quickly, or at all, unaided). " True, but spreadsheets and inventory programs would be of little immediate help in sitting down to an IQ test. In the particular domain of IQ tests, what partitioning of human/machine skills would make sense? For me it comes down to the question of artificial intelligence as a replacement or an assistant to human intelligence. Who would you hire for the job, the computer system with an IQ of 200, a person with an IQ of 150, or a person with an IQ of 125 who knows how to work with a computer system with an IQ of 200 :-)? ------------------------------ Date: Fri, 22 Nov 85 08:10 EST From: D E Stevenson Subject: Hard Typing I asked our local software engineer about investigations using hard typing. His response was that such investigations have been going on since about 1977. Much of this was done by Gannon. Anyone interested in such work can contact David Hutchens on Hutch@clemson. Steve ------------------------------ Date: Thu 21 Nov 85 22:00:11-PST From: Ken Laws Subject: Spang Robinson Report Newsletter news: The TARGET newsletter has merged with The Artificial Intelligence Report to form the Spang Robinson Report. Sara Spang is the editor, Louis G. Robinson the publisher. The report is available on-line through NewsNet and Mead Data Central's Nexis. For subscription info contact Spang Robinson, 3600 West Bayshore Road, Palo Alto, CA 94303. (It's $295 U.S. and Canada, $345 elsewhere.) The contents of the first issue were listed in a recent AIList message from Laurence Leff. ------------------------------ Date: Thu, 21 Nov 85 13:48:10 EST From: Norman Haas Subject: A.I. to the rescue! [Forwarded from the SRI bboard by Laws@SRI-AI.] The following is a slightly abbreviated version of a recent press release. It's interesting what applications AI is being put to these days... Norm Corporate News Bulletin November 19, 1985 IBM ANNOUNCES NEW PROGRAM TO STRUCTURE COBOL CODE IBM today announced a software program that uses "artificial intelligence" techniques to structure existing programs written in the COBOL language so they can be maintained and modified more easily. COBOL Structuring Facility (COBOL/SF) analyzes complex programs, structures them into a top-down hierarchy of components and identifies components for possible re-use in new program development. This can extend the life of current programs, reduce the time and cost to develop new ones and improve the productivity and quality of program maintenance. The "artificial intelligence" techniques used to organize the unstructured program into a hierarchy of components resemble the way a person would analyze and reorganize the program. Since COBOL (Common Business Oriented Language) was one of the first and most widely used computer programming languages, many businesses and organizations depend on COBOL-developed programs that have been modified repeatedly. This can result in programs that are more and more complex and difficult to maintain and can increase the risk of making changes that have hidden consequences. Programmers can use it to structure old code -- helping make programs easier to understand, maintain and modify -- and keep programs in structured form despite continued modifications. It has two modes of operation: Analysis Mode and Generation Mode. In Analysis Mode before structuring, it analyzes the program and produces a report that identifies unreachable code and endless loops it has detected. The report also lists especially complex parts of the program where minor modifications by a programmer could improve the structuring process. In Generation Mode, it structures the program and organizes it in a hierarchy of individually structured procedures. This process -- which resembles the way a person would analyze and reorganize the program -- is an example of what many computer scientists refer to as "artificial intelligence." After a program is structured, a report summarizes the changes by it and provides a structured table of program contents. The structured program then must be recompiled. ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Fri Dec 20 02:12:25 1985 Date: Fri, 20 Dec 85 02:12:05 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a017979; 24 Nov 85 1:14 EST Date: Sat 23 Nov 1985 21:40-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #177 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Sun, 24 Nov 85 05:43 EST AIList Digest Sunday, 24 Nov 1985 Volume 3 : Issue 177 Today's Topics: Query - Connectionist BBoard, Science Fiction - Machines That Talk, Literature - Technical Translation Group, Expert Systems - Liability, Intelligence - Modeling Prejudice ---------------------------------------------------------------------- Date: Sat 23 Nov 85 18:07:06-PST From: Lee Altenberg Subject: connectionist query Does anyone know of a connectionist mailing list or bulletin board to read about the latest stuff in this area? -Lee Altenberg@sumex-aim.arpa ------------------------------ Date: Thu, 21 Nov 85 14:33:59 est From: ulysses!blade!gamma!mike@ucbvax.berkeley.edu (Mike Lukacs) Subject: Machines That Talk Fictional accounts of machines that talk: see: "The Moon is a Harsh Mistress" by Robert A. Heinlein not the earliest account but a good detailed discussion of the (fictional) process by which Adam Selene (a self-aware computer) produced natural language and a video self image. Michael E. Lukacs Bell Communications Research NVC 3X-330 (201)758-2876 ! \ .OR. \ ! \ .OR. > bellcore!sabre!nyquist!maxwell!mike direct ----> ----->-----/ ------------------------------ Date: Fri, 22 Nov 85 10:37:21 pst From: eugene@AMES-NAS.ARPA (Eugene Miya) Subject: Misunderstanding about the technical translation group Recently, several members of the group volunteering technical translation time (including myself) have been asked by people on the net for copies of papers we have identified as significant. PLEASE DO NOT ASK US FOR COPIES, especially translated copies. We are NOT a translation service: this is too expensive of our time. We simply identify (by translating the titles, authors, journals, and subjects for the readership of existing newsgroups) potentially significant papers. We try to provide source material when ever possible (institution or publishing house). Some readers may find this like "Tantalus and the Grapes." We can stop, and you don't ever to hear about these publications again. Just complain. This is an international volunteer effort. I've seen some postings in their native European language (I should have told some people to use English as a common Aether). It is up to the readership of the net to obtain foreign language reports just as they obtain other TRs. --eugene miya NASA Ames Research Center ------------------------------ Date: Fri, 22 Nov 85 18:57 EST From: Stephen G. Rowley Subject: Expert Systems and Liability Date: Wed, 20 Nov 85 12:02:13 est From: mayerk%UPenn-GradEd%upenn.csnet@CSNET-RELAY.ARPA Subject: Issues Concerning Expert Systems -- Who Owns What I think that in the next few years, serious questions like these will affect the way we think about computer systems. Another hard question is who is liable? Let's say some expert system in medical diagnosis was fed data, which is later found to be faulty, or downright wrong. And a serious injury was related to a physician's use of this information. If the writers of the expert system could be shown to be negligent in the verification of their data, could it be possible that they are liable, just as if it were a case of malpractice? The questions get even more interesting than that! Suppose there's a real zippy expert system available to your doctor. He has to decide whether or not to use it, and, if he uses it, whether or not to take its advice. [1] Suppose he decides NOT to use it, and he later messes up in a way the program would have warned him about. Can you sue him for NOT using the latest technology? If not, how is this different from a doctor who refuses to use CAT scans, X-rays, tissue-typing, or anything else? [2] Suppose he decides to use it. [2a] The program tells him to pursue a particular treatment, which he does. You are injured as a result. Can you sue the doctor? Or should you sue the program's sellers? Suppose the program's sellers and implementors are not the same; can you sue them both? What if the doctor "should have known better"? (Never apply a tourniquet about the neck...) [2b] He overrides the program and does something else. You are injured, either as a result of his treatment, or lack of the treatment the program ordered, or both. One consistent interpretation is that the doctor is ALWAYS responsible. [That's why people usually trust a doctor; he's paid to take the responsibility.] On the other hand, you could reasonably claim that it's not his fault if the technology misled him... ------------------------------ Date: 20 Nov 1985 1733-PST (Wednesday) From: aurora!eugene@RIACS.ARPA (Eugene miya) Subject: Re: Removing prejudice (actually question on AI) > Date: Thu, 14 Nov 85 22:32:12 PST > From: Richard K. Jennings > Subject: Removing Prejudice > > I think modeling prejudice accurately is the mandate for AI systems, not > removing it. > > Rich. I think the term you mean to use is "discriminate" in the behaviorism sense. I know the keyword you have in mind is "accurately," but I cannot escape the problem that Intelligence is merely "accurate preudice or discrimination." Recently, I stuck the "artificial" adjective in front of various words: artificial prejudice, artificial discrimination [all this before Rich's posting]. It occurs to me that a lot of what is lacking in AI would be illustrated by the difference of these and what be called "artificial compassion." I am not arguing for emotion in AI, but let me give you the circumstances where I thought of this. Recently, I met with some people who wanted to build an expert system to help with Federal government procurement. It occured to me that the Government might get the idea to build expert systems as social workers to process welfare cases or perhaps scholarship application. That type of work is not merely accessing numbers or conditions (frames?). Deception is obviously a problem. It occurs to me that creating "artificial compassion" in this type of example is much harder than building artificial discriminators. This type of activity is the type which humans pride themselves, and perhaps makes up a portion of "intelligence." Again, take the emotional aspect out of this, and I wonder one might implement the "social worker" system [high human goal]. Subexpert systems to aid overworked humans need not apply. Can we build such a system? Lastly, the above comment falls into the category of social implications of AI. I am uncertain about all the AI issues I have raised, but it seems to me that if corporations get the ideas of making such systems for things like credit ratings [yes, I know you can argue they do this now], or in worse cases to support "evil" governments, then ..... My interest is in the rule-system when "rules" are broken: the welfare worker decides to give assistance or the student gets the scholarship when rule say NO. I'm confused, and all explanations are welcome. Cite Turing's original paper if you would like to show what area I'm getting wrong. --eugene miya NASA Ames Research Center {hplabs,hao,dual,ihnp4,vortex}!ames!amelia!eugene eugene@ames-nas.ARPA ------------------------------ Date: Fri, 22 Nov 85 14:58 EST From: Mukhop Subject: Modeling Prejudice > From: Richard K. Jennings > Subject: Removing Prejudice > Concerning removing prejudice ....... > I think modeling prejudice accurately is the mandate for AI > systems, not removing it. Modeling prejudice accurately is the mandate for AI systems inasmuch as prejudicial inferencing embodies default reasoning and generalizing. Admittedly, these are powerful techniques, but they can sometimes cause unsound inferences. Over-generalization from a small sample size and upward inheritance of defaults (most sports cars are two-seaters => most cars are two-seaters) are error-prone. Modeling those aspects of prejudicial reasoning that cause irrational behavior is certainly not mandated for common-sense reasoning in AI systems. After all, some people exhibit more (robust) common sense than others and it would be worthwhile modeling a clear thinker. The upward inheritance of defaults may actually be used with great advantage, but the clear thinker is aware of its limitations and uses it with caution. My original submission regarding prejudice and rumor addresses the design issues of a higher level control structure for selectively invoking the appropriate reasoning techniques. This requires a good knowledge of the degradation criteria for each technique--an important component of common sense. Uttam Mukhopadhyay Computer Science Dept. GM Research Labs Warren, MI 48090-9055 Phone: (313) 575-2105 ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Fri Dec 20 01:58:28 1985 Date: Fri, 20 Dec 85 01:58:13 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a018351; 24 Nov 85 2:56 EST Date: Sat 23 Nov 1985 21:50-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #178 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Sun, 24 Nov 85 05:44 EST AIList Digest Sunday, 24 Nov 1985 Volume 3 : Issue 178 Today's Topics: Humor - Confectionist Seminar & SARTRE & Advanced Learning Techniques & Proof Methodologies & Rinaldo's laws ---------------------------------------------------------------------- Date: 11/21/85 14:21:07 From: ELISHA at MIT-MC.ARPA Subject: Confectionist Autumn Repast [Forwarded from the MIT bboard by SASW@MIT-MC.] CONFECTIONIST MODELS: AN AUTUMN REPAST Sponsored by the Pavlov Foundation ORGANIZERS: Siu-ling Ku Peter Nuth Margaret St. Pierre Steve Seda DATE: November 22, 1985 PLACE: MIT A.I. Lab, 8th Floor Playroom PURPOSE OF THE PROGRAM: The purpose of the autumn repast is to familiarize young researchers with current techniques in the area of confectionist models of intelligence. This includes search procedures, learning procedures, and methods for representing knowledge in massively parallel networks of carbohydrate units. Application areas include vision, speech, associative memory, natural language and motor control. FACULTY: There will be four full time Tutors plus several Guest Lecturers. Tutors: Guest Lecturers: James Horner Julia Child (Ms) D. Muffet Charles Beard (on videotape) Peter Piper Elizabeth Crocker Thomas Piperson others to be announced [...] ------------------------------ Date: Thu 21 Nov 85 12:34:58-PST From: Paul Roberts Subject: A VALGOL-(un)like language...... [Forwarded from the Stanford bboard by Laws@SRI-AI.] A New Programming Language: SARTRE *SARTRE--Named after the late existential philosopher, SARTRE is an extremely unstructured language. Statements in SARTRE have no purpose; they just are. Thus SARTRE programs are left to define their own functions. SARTRE programmers tend to be boring and depressed and are no fun at parties. The SARTRE language has two basic data types, the EN-SOI and the POUR-SOI. The EN-SOI is a completely filled heap, whereas the POUR-SOI is a dynamic structure which never has the same value. The structures are accessed through the only operation defined in SARTRE, nihilation, which usually results in a ?BAD FAITH at PC 02AC040 error. Comparisons in SARTRE have a peculiar form in that the IF statement can take no arguments and simply reads IF; Similarly, assignments can only be of the form WHAT-IS := (NOT WHAT-IS); since in SARTRE the POUR-SOI is only, and exactly, what it is not. Although this sounds confusing, a background process, the NIHILATOR, is constantly running, making any such statements (or any statements at all, for that matter), completely meaningless. Programs in SARTRE do not terminate, of course, since there is No Exit. --Author Unknown ------------------------------ Date: Mon 11 Nov 85 13:57:56-PST From: Paul Roberts Subject: Advanced Learning Techniques.... [Forwarded from the Stanford bboard by Laws@SRI-AI.] [Disclaimer: all this is forwarded. "I" and "me" don't apply to I or me. -- PR] Columbia is extending its lead the learning field: 3 recent discoveries have just been made by grad students there. LEARNING BY EVOLUTION - ako survival of the fitest. Doesn't work for qualifiers because admission to the department does not extend to the progeny. LEARNING BY SENORY DEPRIVATION - staying away from the department in order to "get some work done", which simply results in (a) having a comprehensive knowledge of the library or (b) having a very clean apartment. LEARNING BY FORGETING EVERYTHING THEY EVER TAUGHT YOU IN SCHOOL - This should be self-explanatory; if not, see me after class. Discoveries about learning itself are a perfect example of "learning by meta-learning". it is well known in ai that if x is good, meta-x is better (or at least meta-x sounds better in the title of a paper). in this case, perhaps meta-learning is symptomatic of learning- avoidance, a common affliction among grad students who prefer to call it "learning-by-doing-anything-else". i know; i've mastered the technique. ------------------------------ Date: 19 Nov 1985 0026-PST (Tuesday) From: Stuart Marks Subject: Found! list of proof methodologies [Forwarded from the Stanford bboard by Laws@SRI-AI.] I have received several responses to my request for proof techniques, some with pointers, and some with actual "proofs." But credit goes to Greg Satz, who dug out of his jokes archive the list that I had in mind. The original author is someone named Dana Angluin, for whom no professional association was given. [John McCarthy reports that Dana Angluin is now in the Computer Science Department at Yale, but probably compiled this list while a graduate student at UCB. -- KIL] There were a couple of references to the following work: Dunmore, Paul V., "The Uses of Fallacy", in R. L. Weber, @i{A Random Walk in Science}. New York: Crane, Russak, & Co. Inc., 1973, p. 29. This contains a similar list of proof techniques. I haven't looked it up yet, but I'll report if I find anything of interest. Here is Dana Angluin's list. ======================================================================= Proof by example: The author gives only the case n=2 and suggests that it contains most of the ideas of the general proof. Proof by intimidation: 'Trivial.' Proof by vigorous handwaving: Works well in a classroom or seminar setting. Proof by cumbersome notation: Best done with access to at least four alphabets and special symbols. Proof by exhaustion: An issue or two of a journal devoted to your proof is useful. Proof by omission: 'The reader may easily supply the details.' 'The other 253 cases are analogous.' '...' Proof by obfuscation: A long plotless sequence of true and/or meaningless syntactically related statements. Proof by wishful citation: The author cites the negation, converse, or generalization of a theorem from the literature to support his claims. Proof by funding: How could three different government agencies be wrong? Proof by eminent authority: 'I saw Karp in the elevator and he said it was probably NP-complete.' Proof by personal communication: 'Eight-dimensional colored cycle stripping is NP-complete' [Karp, personal commmunication]. Proof by reduction to the wrong problem: 'To see that infinite-dimensional colored cycle stripping is decidable, we reduce it to the halting problem.' Proof by reference to inaccessible literature: The author cites a simple corollary of a theorem to be found in a privately circulated memoir of the Slovenian Philological Society, 1883. Proof by importance: A large body of useful consequences all follow from the proposition in question. Proof by accumulated evidence: Long and diligent search has not revealed a counterexample. Proof by cosmology: The negation of the proposition is unimaginable or meaningless. Popular for proofs of the existence of God. Proof by mutual reference: In reference A, Theorem 5 is said to follow from Theorem 3 in reference B, which is shown to follow from Corollary 6.2 in reference C, which is an easy consequence of Theorem 5 in reference A. Proof by metaproof: A method is given to construct the desired proof. The correctness of the method is proved by any of these techniques. Proof by picture: A more convincing form of proof by example. Combines well with proof by omission. Proof by vehement assertion: It is useful to have some kind of authority relation to the audience. Proof by ghost reference: Nothing even remotely resembling the cited theorem appears in the reference given. Proof by forward reference: Reference is usually to a forthcoming paper of the author, which is often not as forthcoming as at first. Proof by semantic shift: Some standard but inconvenient definitions are changed for the statement of the result. Proof by appeal to intuition: Cloud-shaped drawings frequently help here. ------------------------------ Date: 20 Nov 85 2304 PST Don Woods Subject: Proof Methodologies [Forwarded from the Stanford bboard by Laws@SRI-AI.] The list of proof methodologies also appeared in SIGACT News, v15 #1 (Spring '83). Incidentally, it omits the one I first heard from RPG, who suggested the following as the generic form of proof methodology used in some theological argument or other: Proof by elimination of the counterexample: 'Assume for the moment that the hypothesis is true. Now, let's suppose we find a counterexample. So what? QED.' ------------------------------ Date: 22 Nov 1985 9:37-PST From: Soon Yau Kong Subject: addendum to list of proof methodologies [Forwarded from the Stanford bboard by Laws@SRI-AI.] "For the last century no one acquainted with the facts has disputed ... - An equivalent statement is "I didn't look up the actual facts but since most people I know think this way,it follows that everyone else does too". Also called proof by assumption -Soon [This was in reference to a bboard discussion on evolution. -- KIL] ------------------------------ Date: Thu 14 Nov 85 02:24:47-PST From: William "Chops" Westfield Subject: Humor: Rinaldo's laws [Forwarded from the Stanford bboard by Laws@SRI-AI.] ***** sri-unix:net.ham-radio / eagle!karn / 4:17 pm Apr 18, 1983 The following was written by Paul Rinaldo, W4RI, until recently the President of AMRAD (the Amateur Research and Development Corp, a Washington-area organization of experimentally-minded radio amateurs.) Paul has accepted an opportunity to run the Technical Department of the ARRL. Rinaldo's Laws As I will be leaving the Washington area in early May, I thought it appropriate to share the wisdom that I have accumulated thus far. These truths have come not as a vision but by observation over time. Accordingly, I have synthesized the following laws: First Law. Choreography is its own reward. Some things are done only for the sake of form. Don't fight it by looking for substance in everything. Do it long enough and you'll find enjoyment in an elephant dance. Second Law. He who does the work shapes it. As applied to computers, he who writes the code rules (the Codin' rule). In meetings, he who writes the minutes determines the outcome. Third Law. The less the knowledge, the more jealously it is preserved. Societies with only a few precious facts make their people memorize them and pledge to faithfully abide by them. In contrast, highly developed disciplines quit worrying about losing knowledge (unless the computer crashes and there is no backup). Fourth Law. Excellence increases demands. Critics gather to spot tinier flaws as work nears perfection. Promptness invites impatience. In correspondence, the faster you answer a letter, the faster your correspondent will answer giving you something with a shorter deadline. This reaches a fever pitch with electronic mail. Fifth Law. Skills diminish professionalism. Engineers who admit to drafting skills are vulnerable to assignment of drafting work, just to help out. Similarly, female professionals should hide any clerical skills lest they be asked to pinch hit for one of the secretaries in the event of illness. Sixth Law. What separates the competent from the incompetent is the ability to cover up mistakes. Many successful sales demonstrations have been made with defective products in the hands of competent persons who avoid demonstrating the features which don't work. Beautiful Xerox copies can be made from originals riddled with correction fluid. Recovery from some grievous errors can be attained by simply announcing, "No problem. We'll just put it back in the word processor!" The computer software profession seems to be the exception; who else is so blatant as to have a term such as "debugging" to let the world know that they need extra time funded by the customer to correct their own errors. Seventh Law. Silence is not acquiescence. Contrary to what you may have heard, silence of those present is not necessarily consent, even the reluctant variety. They simply may sit in stunned silence and figure ways of sabotaging the plan after they regain their composure. Eighth Law. Quick-reaction and slow-reaction facilities rotate. Once people discover that there is a quick-reaction facility (QRF), they will try to get all their work done there, bogging it down in work and leaving the slow-reaction facility (SRF) nothing to do, thus becoming the faster of the two. Ninth Law. Complexity attracts brilliance. The KISS (keep it simple, stupid) principle is no fun and certainly not a professional approach. If you want brilliant people to do work for you make it complex and demanding. The true professional will spend 20 hours at the computer writing a one-time-use program that will replace 10 hours of clerical work. Anyway, 20 hours at professional rates pays more than 10 hours at clerical rates. Also, it's more intellectually rewarding. The greatest achievement is to use one's finest professional talents to accomplish something that didn't need to be done. Tenth Law. Bad guys are replaced. Did you ever rejoice over the departure of someone that you couldn't get along with only to find that a replica has shown up? When you are trying to make a U-turn and you have someone tailgating you, have you pulled off on a sidestreet, then into an alley only to find that two other cars are right behind you? ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Fri Dec 20 01:52:42 1985 Date: Fri, 20 Dec 85 01:52:16 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a001766; 25 Nov 85 14:00 EST Date: Mon 25 Nov 1985 09:26-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #179 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Mon, 25 Nov 85 22:47 EST AIList Digest Monday, 25 Nov 1985 Volume 3 : Issue 179 Today's Topics: Literature - Recent Articles ---------------------------------------------------------------------- Date: 24 Nov 1985 19:55-CST From: leff%smu.csnet@CSNET-RELAY.ARPA Subject: Recent Articles %A Christine McGeever %T Databases: Artificial Intelligence Breeds Competition for Established Vendors %J Infoworld %V 7 %N 43 %D OCT 28, 1985 %K Paradox Symantec Q&A ANSA micro database %X Describes two database management systems which run on IBM PC's and which are allegedly using Artificial Intelligence. Paradox is a system which uses Query by Example with an Artifical Intelligence system that optimizes the query. Q&A uses an English front end and also uses query by example. %A Keith Thompson %T Paradox: Powerful Pricey, and Easy %J Infoworld %V 7 %N 43 %D OCT 28, 1985 %K Database query by example micro %X review of the Paradox system which uses artificial intelligence to optimize queries. It earned three diskettes out of five and had the following ratings performance: very good documentation: satisfactory ease of learning: very good ease of use: very good error handling: poor support: satisfactory value: satisfactory %A Dick Pountain %T Computers as Consultants %J BYTE %P 367-376 %V 10 %N 10 %D OCT 1985 %K Expert-Ease Tess EMYCIN micro Expert Edge %X reviews expert-ease (a decision tree generating system) and Tess which is an EMYCIN like program. %A Kenneth W. Kerber %T Review of Computer Culture: The Scientific, Intellectual and Social Impact of the Computer %J BYTE %P 57-58 %V 10 %N 10 %D OCT 1985 %X This book includes papers by John McCarthy (on the need for commonsense in expert sytems), Pamela McCorduck (on the promise of fifth generation computers), Seymour Papert (on cognitive reasoning), Daniel Dennett (on using computers to extend senses and imagination) and Feigenbaum (on automated knowledge acquisiton in expert systems) %A B. W. Wah %A C. F. Yu %T Stochastic Modeling of Branch-and-Bound Algorithms with Best-First Search %J IEEE Transactions on Software Engineering %V SE-11 %N 9 %D SEP 1985 %P 922-923 %X reprint of paper from Compsac 82 %A B. P. McCune %A R. M. Tong %A J. S. Dean %A D. G. Shapiro %T RUBRIC: A System for Rule-Based Information Retrieval %J IEEE Transactions on Software Engineering %V SE-11 %N 9 %D SEP 1985 %P 939-944 %A D. Keirsey %A J. Mitchell %A B. Bullock %A T. Nussmeier %A D. Y. Tseng %T Autonomous Vehicle Control Using AI Techniques %J IEEE Transactions on Software Engineering %D SEP 1985 %V SE-11 %N 9 %P 9086-991 %A Richard Fikes %A Tom Kehler %T The Role of Frame Based Representation in Reasoning %J CACM %D SEP 1985 %V 28 %N 9 %P 904-920 %K STAR_PLAN satellite alarms KEE %X describes STAR-PLAN which uses alarms to wake certain parts of an expert system when conditiosn requires its analysis. Also discusses frames in general and KEE. %A Frederick Hayes-Roth %T Rule-Based Systems %J CACM %D SEP 1985 %V 28 %N 9 %P 921-932 %A Michael R. Genesereth %A Matthew L. Ginsbert %T Logic Programming %J CACM %D SEP 1985 %V 28 %N 9 %P 933-941 %X prolog %A James R. Slagle %A Henry Hamburger %T An Expert System for a Resource Allocation Problem %J CACM %D SEP 1985 %V 28 %N 9 %P 994-1004 %K Battle %X describes a system for deciding how to allocate weapons. Goes into detail on its variation of A* to come up with a good approximation to the optimum configuration and how values are propagated through the network. The techniques described could be used in such applications as allocating advertising time in dealing with competitors in the marketplace. %A S. D. Morgera %T Information Theoretic Covariance Complexity and its Relation to Pattern Recognition %J IEEE Transactions on Systems, Man and Cybernetics %D SEP/OCT 1985 %V SMC-15 %N 5 %P 608-619 %A A. Goshtasby %A G. C. Stockman %T Point Pattern Matching Using Convex Hull Edges %J IEEE Transactions on Systems, Man and Cybernetics %D SEP/OCT 1985 %V SMC-15 %N 5 %P 631-636 %A R. Jakubowski %T Extraction of Shape Features for Syntactic Recongition of Mechanical Parts %J IEEE Transactions on Systems, Man and Cybernetics %D SEP/OCT 1985 %V SMC-15 %N 5 %P 642-651 %T See $980M Sales in Vision Systems in 90 %J Electronic News %D SEP 16, 1985 %V 31 %N 1567 %P 21 %X Frost and Sullivan predicts a 980 million market in vision systems in 1990 with 480 million of which would be inspection and 200 million for identification and sorting %A Paul J. Besl %A Ramesh C. Jain %T Three-Dimensional Object Recognition %J ACM Computing Surveys %V 17 %N 1 %D MAR 1985 %P 75-145 %T New Configuration Added, Price Cut, For Processor %J Computer %P 122 %D MAY 1985 %V 18 %N 5 %K Symbolics lisp machine %X Symbolics announced a new configuration of its 3640 processor, the 3640-1711 which has two 140 MB Winchester disks and costs $71,080 and with one 140MB hard drive it costs $65,900. 2MB add-in board is $9,900 and an eight megabyte add-in memory board is $35, 000. %T TI Offers Personal Consultant Version %J ComputerWorld %D SEP 16, 1985 %V 19 %N 37 %P 61 %K micro %X discusses the new Personal Consultant Plus at $2950 and the new price for the old Personal Consultant of $950.00 %T Microcomputers %J ComputerWorld %D SEP 16, 1985 %V 19 %N 37 %P 66 %K RuleMaster Radian micro %X discusses enhancements to its Rulemaster with new screen and menu handling facilities. %A Takehisa Kondoh %T Japanese Consortium Set %J ComputerWorld %D SEP 16, 1985 %V 19 %N 37 %P 85 %K Nissin Structural Planning Institute JGC Intelligence Engineering Research Institute German Center for Comptuer Science %X describes a consortium called Intelligence Engineering Research Institute to be formed by European, US and Japanese companies. %T Insight to Interpret with Expert Reports on Mac %J Infoworld %V 7 %N 37 %D SEP 16, 1985 %P 61 %K Macintosh micro Layered Inc. accounting %X describes an accounting system with an AI program to help interpret the reports %T A Device that Heads Off Road Collision %J Business Week %D SEP 2, 1985 %P 77 %K Vehicle Radar Safety Systems, Inc. %X Describes a system to warn drivers of impending collisons. It does not falsely alarm for road signs and radar overpasses. %T Frey Associations to Exhibit Enhanced Version of Themis %J DEC HARDCOPY %V 14 %N 5 %P 94 %D MAY 1985 %K natural language database %X this is one of those systems that allow the user to give his queries in English to the database. It supports 1000 word vocabulary which can be expanded by the user. It works with Oracle's Relational Database Managmenet System and VaX Datatrieve. It sells for $24,000. %T Natural Language Software Builds Mainframe Databases %J ElectronicsWeek %D JAN 7, 1985 %V 58 %N 2 %P 49 %K Artificial Intelligence Corporation Intellect %X They announced Intellect/SX which allows users to set up their own lexicons, databases and manipulate their own databases. %T Bureaucracy Spawns Entrepreneurs %J High Technology %D FEB 1985 %P 20 %V 5 %N 2 %K Lockheed pilot airplain Getex navigation %X Describers AI system to airplane navigation making it possible to have a one pilot airplane. %A C. P. Neuman %A V. D. Tourassis %T Discrete Dynamic Robot Models %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 193-203 %A J. D. Wolter %A R. A. Volz %A Anthony C. Woo %T Automatic Generation of Gripping Positions %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 204-212 %A B. K. Kim %A K. G. Shin %T Minimum-Time Path Planning for Robot Arms and Their Dynamics %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 213-223 %A R. A. Brooks %A T. Lozano-Perez %T A Subdividsion Algorithm in Configuration Space for Findpath with Rotation %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 224-233 %A J. Rasmussen %T The Role of Hierarchical Knowledge Represenation in Decisionmaking and System Management %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 223-243 %A Mark R. Laff %A Brent Halpert %T SW 2 - An Object-based Programming Environment %J SIGPLAN 85 Symposium on Language Issues in Programming Environments %C Seattle, Washingto/n %D JUN 25-28, 1985 %I ACM %P 1-11 %A James Purtilo %T Polylith: An Environment to Support Management of Tool Interfaces %J SIGPLAN 85 Symposium on Language Issues in Programming Environments %C Seattle, Washingto/n %D JUN 25-28, 1985 %I ACM %P 12-18 %K Macsyma %X Describes a front end for Macsyma to allow interfacing it with various other tools as well as general concepts for front-ends. %A Nissim Francez %A Shalom Goldenberg %A Ron Y. Pinter %A Michael Tiomkin %A Shalom Tsur %T An Environment for Logic Program %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 179-190 %A Mark Moriconi %A Dwight F. Hare %T PegaSys: A System for Graphical Explanation of Program Designs %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 148-160 %A Henryk Jan Komorowski %A Shigeo Omori %T A Model and an Implementation of a Logic Programming Environment %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 191-198 %A Stephen Fickas %T Design Issues in a Rule-Based System %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 2 %D MAR/APR 1985 %P 208-215 %A P. E. Lehner %A M. A. Probus %A M. E. Donnell %T Building Decison Aids: Exploiting the Synergy Between Decison Analysis and Artificial Intelligence %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 4 %D JUL/AUG 1985 %P 469-474 %A Y. J. Tejwani %A R. A. Jones %T Machine Recognition of Partial Shapes Using Feature Vectors %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 4 %D JUL/AUG 1985 %P 504-516 %A A. L. Porter %A F. A. Rossini %A J. Eshelman-Bell %A D. Jenkins %A D. J. Cancelleri %T Industrial Robots - A Strategic Forecast Using the Technological Delivery System Approach %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 4 %D JUL/AUG 1985 %P 521-526 %A S. B. Ahuja %A J. A. Reggia %T Automated Classification of Complex Errors in Discrete Sequential Tasks %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 4 %D JUL/AUG 1985 %P 527-531 %A S. Gaglio %A R. Minciardi %A P. P. Puliafto %T Multiperson Decision Aspects in the Construction of Expert Systems %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 4 %D JUL/AUG 1985 %P 536-539 %A Y. Ichikawa %A N. Ozaki %T A Heuristic Planner and an Executive for Mobile Robot Control %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 4 %D JUL/AUG 1985 %P 558-563 %A Charles L. Cohen %T AI Comes to the Aid of Pencil Pushers %J Electronics %P 58 %N 14 %D AUG 26, 1985 %P 18-19 %K pattern recognition VLSI Toshiba Expert System %X Describes an expert system that transforms pencil sketches of integrated circuit masks to computer compatible sketches. This was developed by Toshiba Corporation. It uses 2000 rules and a structure-matching method using model-based segmentation and hypothesis verfication. %A Alexander Wolfe %T How NASA Will Use AI in Space %J Electronics %D SEP 16, 1985 %P 32-327 %V 58 %N 37 %K expert systems space shuttle space station Empress LOX %K AI applications at NASA management of space resources controlling experiments, assisting in repairng planning and scheduling system for payload processing diagnostic system for loading liquid oxygen into the shuttle scheduling the operations of the tracking and data-relay satellite %A James Clifford %A David S. Warren %T Formal Semantics for Time in Databases %J ACM Transactions on Database Systems %D JUN 1983 %V 8 %N 2 %P 214-254 %A R. Emeett Carlyle %T Can AI Save Cobol %J Datamation %D SEP 15 %D SEP 15, 1985 %V 31 %N 18 %K expert systems T. Capers Jones maintenance %X T. Capers Jones has developed expert systems that will structure COBOL that produce code that takes 25 to 50 percent less time to maintain. %A Neil C. Rowe %T Top-down Statistical Estimatation on a Database %J SIGMOD Record %V 13 %N 4 %P 135-145 %K expert systems %X This system uses a set of general-purpose statistics on a database and then a set of inference rules to estimate other arbitrary statistics requested by users %A Janet L. Koldoner %T Indexing and Retrieval Strategies for Natural Language Fact Retrieval %J ACM Transactions on Database Systems %V 8 %N 3 %P 434-464 %D SEP 1983 %K Cyrus information retrieval %T Xerox Adds Low-End Workstations %J Electronic News %V 31 %N 1566 %D SEP 9, 1985 %P 37 %K lisp machines %X Describes new 1185 and 1186 with prices form 10,000 to 19,ppp. The 10,000 gets you an 1185 with 1.1 Mbytes RAM, 10 Mbyte Winchester and a 15 inch display. The 19 grand gets you 3.6M bytes RAM, 80 Mbyte disk and 19 inch display. %T Tektronix Unveils 2 New AI Systems %J Electronic News %V 31 %N 1566 %P 39 %T Kurzweil Names Financial Chief %J Electronic News %V 31 %N 1566 %P 39 %K Bernard F. Bradstreet %X Kurzweill Applied Intelligence named Bernard F. Bradstreet, a former Prime Computer vice president and treasurer, as vice president of finance and administration. %T Thoughtware Set To Buy Lightyear %J Electronic News %V 31 %N 1566 %D SEP 9, 1985 %P 52 %X Thoughtware, a delvoloper of decision support systems has agreed to acquire Lightyear, Inc. %T Raster, TI Sign Marketing PACT %J Electronic News %V 31 %N 1566 %D SEP 9, 1985 %P 42 %K Lisp Machine Graphics Explorer %X Raster Technologies has agreed to market its color graphics system for TI's Explorers. %T Expert System Language May Improve 1-2-3 %J InfoWorld %D NOV 4, 1985 %V 7 %N 44 %P 3 %K Teknowledge Lotus user interface expert system error correction M.1 %X Teknowledge is developing an M.1 front end to Lotus 1-2-3 which will distinguish semantic errors from syntax errors and to allow the user to express logic relationships. It also provides a query by example front end. M.1 is being rewritten in the C programming language. %T Advertisement %J InfoWorld %D NOV 4, 1985 %V 7 %N 44 %P 28 %X Advertisement for AI:Typist which is a $79.00 word processing system that has a spelling corrector that checks words as they are entered. Billed as "AIRUS-A Technology is so new it's still baffling the experts." %T Lisp Machines Adds 2 New Directors %J Electronic News %V 31 %N 1576 %D NOV 18, 1985 %P 27 %X Lisp Machines has added two new people to its board of directors: George J. Demos and Harold Shattuck. Art Marks has replaced Neil Bond on the board. %T Robotic Vision Taps Marketing V-P %J Electronic News %V 31 %N 1576 %D NOV 18, 1985 %P 90 %K Robotic Vision Systems %X Robotic Vision Systems Systems reported a loss for the second quarter of $319,000 compared with a $299,00 loss in the 1984 period. The sales for the three months was 2.7 million up from 1.3 million. Robotic Visions sell three-dimension vision-based robotic systems. They also named L. Knox Johnstone, formerly marketing manager of Robot Defense Systems, as vice president of marketing. %T System for Automated Troubleshooting %J NASA Tech Briefs %D Winter 1985 %V 9 %N 4 %P 166-167 %K maintenance expert system %X Describes software done by Leonard Friedman of Caltech in the area of diagnostic system for electromechanical mechanisms %T Automatic Guidance for Remote Manipulator %J NASA Tech Briefs %D Winter 1985 %V 9 %N 4 %P 68-69 %K robotics %X System based on maintaining a mirror on the object to be grasped with detectors and light sources on the end effector of the robot. It is patented (NO. 3,888,362) %A James A. Marti.n %T Nixdorf Expert System Shell Out %J ComputerWorld %D NOV 4, 1985 %V 19 %N 44 %P 31 %K Prolog MProlog Twaice Logicware %X Nixdorf announced an expert system shell based on Logicware's Prolog. The system supports a taxonomy model, rule compiler and a general lease is $70,000. %T News from the Industry Front %J ComputerWorld %D NOV 4, 1985 %V 19 %N 44 %P 31 %K Inference Carnegie Group Ford Motor %X Ford Motor purchased interest in Inference Corp and Carnegie Group.. These represent $14 million each. %A T. M. Mitchell %A L. I. Steinberg %A J. S. Shulman %T A Knowledge-Based Approach to Design %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 502-510 %K VLSI Design %A G. Kahn %A S. Nowlan %A J. McDermott %T Strategies for Knowledge Acquisition %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 511-522 %K maintenance MUD oil extraction MORE %X describes a general theory for when a domain is suited for diagnostic expert system based on shallow knowledge and how to construct the base. Then this was automated in a tool called MORE %A R. Rada %T Gradualness Facilitates Knowledge Refinement %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 523-530 %X describes a method to determine the weights of rules in expert systems based on the closeness of answers of the expert systems on the given answers %A A. Sathi %A M. S. Fox %A M. Greenberg %T Representaton of Activity Knowledge for Project Management %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 531-552 %A P. K. Fink %A J. C. Lusth %A J. W. Duran %T A General Expert System Design for Diagnostic Problem Solving %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 553-560 %X describes a structure to which any domain that needs to be diagnosed can be described as. This is for diagnostic system based on deep reasoning. %A P. S. Rosenbloom %A J. E. Laird %A J. McDermott %A A. Newell %A E. Orciuch %T R1-Soar: An Experiment in Knowledge-Intensive Programming in a Problem-Solving Architecture %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 561-569 %A D. M. McKeown, Jr. %A W. A. Harvey, Jr. %A J. McDermott %T Rule-Based Interpretation of Aerial Imagery %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 570-585 %K vision %A C. V. Apte %A S. M. Weiss %T An Approach to Expert Control of Interactive Software Systems %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 586-591 %K well-log geology ELAS %X describes a system for dynamic selection and interpretation fo appropriate programs and parameters in the area of well-log analysis %A Yizong Cheng %A King-Sun Fu %T Conceptual Clustering in Knowledge Organization %J IEEE Transactions on Pattern Analysis and Machine Intelligence %V PAMI-7 %N 5 %D SEP 1985 %P 592-598 %K Huato traditional Chinese medicine %X describes an expert system for "traditional Chinese medicine." ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Fri Dec 20 02:10:56 1985 Date: Fri, 20 Dec 85 02:10:26 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a014386; 27 Nov 85 20:11 EST Date: Wed 27 Nov 1985 15:44-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #180 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Tue, 3 Dec 85 23:45 EST AIList Digest Thursday, 28 Nov 1985 Volume 3 : Issue 180 Today's Topics: Queries - AI Workstations & Expert System on COBOL, Science Fiction - Machines That Talk, Humor - Doonesbury on Artificial Intelligence, Expert Systems - Liability & ES Strategies Newsletter, Literature - DNA Analysis & Recent Tech Reports, Seminar - The Riddle of STRIPS (SU) ---------------------------------------------------------------------- Date: 26 NOV 85 12:08-N From: MANUEL%CGEUGE52.BITNET@WISCVM.ARPA Subject: AI Workstations We are looking for an AI workstation for research purposes and would appreciate any comments that you have concerning your favorite (or not so favorite) workstation. In particular, Sun vs. Symbolics arguments would be appreciated. Here are some of the workstations that we are considering: Apollo DOMAIN family Symbolics 3600 DEC VAXstation II Tektronix TEK 4406 LMI Lambda family TI Explorer Sun 2 & 3 Xerox 1100 series Please send your comments directly to me. If you would like the contest results posted then please let me know. Thanks very much for your consideration. ... James Stewart Department of Physics University of Geneva MANUEL @ CGEUGE52 (on BITNET) ------------------------------ Date: Mon 25 Nov 85 13:10:38-EST From: Marc Vilain Subject: Expert system defeats halting problem? I'd be curious to know just which kinds of infinite loops are detected by IBM's COBOL restructuring program. No doubt that some cases can be detected by rudimentary kinds of template matching, but in general what we're talking about is the halting problem. Anybody out there have more information on this? marc. ------------------------------ Date: Sat, 23 Nov 85 10:59:02 GMT From: gcj%qmc-ori.uucp@ucl-cs.arpa Subject: Re: Fictional accounts of machines that talk (Vol 3 # 172) The natural language capabilities of a machine are amply illustrated in Robert Heinlein's ``The Moon is a Harsh Mistress'', (cf AIList Vol 3 # 39), written in the sixties. Here the machine, called Mike, talks to the programmer who realizes that Mike is now large enough to be self-aware. Not only does Mike hold quite normal conversations, he eventually creates a fictional character ``inside his head'', called Adam Selene. He then gives this character life by building a video and voice image. Many people would like to meet Mr. Selene in person! Also worth a mention in this context is ``Do Androids Dream of Electric Sheep?'', by Philip K. Dick. These are not quite the ancient Greeks, or particularly early. Gordon Joly gcj%qmc-ori@ucl-cs.arpa ------------------------------ Date: Sun 24 Nov 85 17:14:02-CST From: Werner Uhrig Subject: Doonesbury on Artificial Intelligence [ Doonesbury - Sunday Nov 24, 1985 ] [ Mike and Bernie, in front of a desk with a MACish computer facing them ] B: You see Mike, most of the computer breakthroughs we're making today have to do with Artificial Intelligence. The computer recognizes patterns in the user's behavior that enable it to make decisions on its own. Of course, the machine is still only as ethical as its owner. Let me show you a hypothetical model. Say I called up my company spread- sheet and transferred $100,000 to my personal account, okay? Watch. >BING!< EMBEZZLEMENT TRANSACTION COMPLETED. M: Good Lord !!! B: Mindless compliance, right? ... RESERVATIONS FOR RIO CONFIRMED. B: ... but it anticipated my every need ! ------------------------------ Date: Sun 24 Nov 85 12:30:41-PST From: PEREIRA@SRI-AI.ARPA Subject: Re: Expert Systems and Liability Given current litigation practices, I would think the answer to your questions is rather simple: you sue everyone in sight (doctor, program seller, program developer, hospital, city, state, Santa Klaus,...). With the ``joint and several'' legal doctrine currently used in liability cases, at least one of them will end up paying... (unless all go bankrupt, of course...) -- Fernando ------------------------------ Date: Wed 26 Nov 85 12:30:41-PST From: Laws@SRI-AI.ARPA Subject: Expert Systems Strategies Newsletter I have received a flyer for another AI newletter, Expert Systems Strategies. Paul Harmon is one of the editors, and his AI in Business book (with David King) is offered free to charter subscribers. The newsletter covers the usual mix: market trends, user profiles, vendor critiques, systems application analyses, performance engineering, hardware reviews, languages and tools, survey reports, knowledge engineering, events calendar, show updates, literature reviews, major contract announcements, and new business ventures -- all focussed on expert systems rather than AI in general. Current price is $207 ($40 off) for 12 monthly issues of 16 pages each. Cahners Newsletter Center, P.O. Box 59, New Town Branch, Boston, MA 02258; (617) 964-3030; Cable/CAHNERS BOSTON; Telex 94-0573 CPC BSN. ------------------------------ Date: Mon 25 Nov 85 01:23:34-CST From: Werner Uhrig Subject: interesting article in CACM of 85/11 1164 DISCOVERING THE SECRETS OF DNA. Peter Friedland and Laurence H Kedes Symbolic Pattern Recognition and the AI methodologies model building and theory formation will be critical to the next stage of discovery in regulatory molecular genetics. ------------------------------ Date: 24 Nov 1985 19:55-CST From: leff%smu.csnet@CSNET-RELAY.ARPA Subject: Recent Tech Reports ADDRESSES TO REQUEST TECH REPORTS LISTED: Department of Computer Science, Oregon State University, Corvallis, OR 97331 Department of Computer Science, Univeristy of Illinois at Urbana-Champign 1304 West Springfield Avenue, Urbana, Illinois 61801 OR erna@uiuc OR Engineering Docurments Center, University of Illinois at Urbana-Champaign, 1308 West Green Street, Urbana, Illinois 61801 Washington State University, Computer Science Deparmtent, Pullman, Washington 99164-1210 Computing Research Laboratory, University of Michigan, Room 1079, East Engineering Building, Ann Arbor, Michigan, 48109, USA UCLA Computer Science Department, 3732 Boelter Hall, Los Angeles, CA 90024 %A Asya Campbell %T Comparison of Dynamically and Statically Scoped Lisp %R CSD-850024 %I UCLA Computer Science Department %X $2.75 %A Rina Dechhter %T Studies in the Use and Generation of Heuristics %R CSD-850033 %I UCLA Computer Science Department %X $12.50 (Ph. D. Thesis) %K A* %A Judea Pearl %T Bayesian Networks: A Model of Self-Activiated Memory for Evidential Reasoning %R CSD-850021 %I UCLA Computer Science Department %X $0.50 %A Judea Pearl %T A Constraint-Propagation Approach to Probabilistic Reasoning %R CSD-850020 %I UCLA Computer Science Department %X $0.75 %A Judea Pearl %T Fusion, Propagation, and Structuring in Bayesian Networks %R CSD-850022 %I UCLA Computer Science Department %X $4.25 %A Judea Pearl %T Bayes Decision Methods %R CSD-850023 %I UCLA Computer Science Department %X $2.00 %A Judea Pearl %A Michael Taria %T Structuring Causal Trees %R CSD-850029 %I UCLA Computer Science Department %X $1.75 %A Judea Pearl %T How to Do With Probabilities What People Say You Can't %R CSD-850031 %X $1.75 %I UCLA Computer Science Department %X $1.75 %A Judea Pearl %T On Evidential Reasoning in a Hierarchy of Hypotheses %R CSD-850032 %X $0.75 %I UCLA Computer Science Department %A Michael A. Langston %A Chul E. Kim %T Movement Coordination for Single-Track Robot Systems %R CSD-84-125 %X $2.50 (discusses dealing with multiple robots) %I Washington State University Computer Science Department %A Jerzy Tiuryn %T An Introduction to First-Order Programming Logics %I Washington State University Computer Science Department %R CSD-84-126 %X $4.20 %A Keshav Sharma %T Syntactic Aspects of the Non-Deterministic Lambda Calculus %R CSD-84-127 %I Washington State University Computer Science Department %X $7.20 %A Jerzy Tiuryn %T A Simplified Proof of DDL < DL %R CS-85-130 %I Washington State University Computer Science Department %X 1.50 (compares Deterministic Dynamic Logics and Dynamic Logics of regular programs) %A David Matthew Dahlbacka %T An ATN-Based Restricted Natural Language Front End for A Data-Flow Design Aid %R Department of Computer Science File No. 944 %I University of Illinois at Urbana-Champaign %D JUL 1985 %K software engineering %A Tomoyasu-Taguti Nakagawa %A Hung-Chi Lai %T Reference Manual of Fortran Program ILLOD-(NOR-B) for Optimal NOR Networks %R Department of Computer Science Report No. 1129 %I University of Illinois at Urbana-Champaign %D JUL 1985 %K branch and bound %A Simon M. Kaplan %T Verification of Recursive Programs: A Temporal Proof Approach %R Department of Computer Science Report NO. 1207 %I University of Illinois at Urbana-Champaign %D SEP 1985 %A Larry Rendell %T Induction, Of and By Probability %R Department of Computer Science Report NO. 1209 %I University of Illinois %D JUL 1985 %X This paper examines some methods and ideas underlying the author's successful probabilistic learning systems (PLS). These systems have proven uniquely effective and efficent for generalization learning (induction) in heuristic search. Aspects of PLS include use of probabilities to guide both task performance and learning, incremental revision and normalization of probabilities, and localization and correction of their errors. Construction of new terms (features) for heuristic functions may be feasible. %A Larry Rendell %T Genetic Plans and the Probabilistic Learning System: Synthesis and Results %R Department of Computer Science Report NO. 1217 %I University of Illinois at Urbana-Champaign %D JUL 1985 %X describes PLS2 which clusters data into economical cells in augmented feature space, and a genetic level which selects successful regions by a genetic algorithm. %A Nachum Dershowitz %T Termination of Rewriting %R Department of Computer Science Report NO. 1220 %I University of Illinois at Urbana-Champaign %D AUG 1985 %A Jean-Luc Remy %T Review of Several Closure Properties in Universal Algebra and First Order Logic %R Department of Computer Science Report NO. 1221 %I Univeristy of Illinois at Urbana-Champaign %D JUL 1985 %A John Shilling %T Initial Report on ISADORE: A Reference Librarian Generator %R Department of Computer Science Report NO. 1225 %I University of Illinois at Urbana-Champaign %D SEP 1985 %A S. R. Ray %A W. D. Lee %A C. D. Morgan %A W. Airth-Kindree %T Computer Sleep Stage Scoring - An Expert System Approach %R Department of Computer Science Report No. 1228 %I University of Illinois at Urbana-Champaign %D September 1985 %A T. G. Lewwis %T An Operator Calculus for Computer Programs %R CSTR-78-1-5 %I Oregon State University, Department of Computer Science %D 1978 %A W. S. Bregar %A A. M. Farley %T Interactive Problem Solving in Elementary Algebra %R CSTR-78-3-1 %I Oregon State University, Department of Computer Science %D 1978 %A P. Cull %A W. Frank %T Flaws of Form %D 1978 %R CSTR-78-20-3 %I Oregon State University, Department of Computer Science %X G. Spencer Brown's book \fBLaws of Form\fR has been enjoying a vogue among social and biological scientists. Proponents claim that the book introduces a new logic ideally suited to their fields of study, and that the new logic solves the problems of self-reference. These claims are false. We show that Brown's system is Boolean algebra in an obscure notation, and that his "solutions" to the problems of self-reference are based on a misunderstanding of Russell's paradox. %A B. Levin %T The Automated Inference of Tree Systems %D 1979 %R CSTR-79-20-6 %I Oregon State University, Department of Computer Science %A B. Blanchard %A W. S. Bregar %T A Production Environment %D 1982 %R CSTR-82-3-1 %I Oregon State University, Department of Computer Science %A D. Kogan %A M. J. Freiling %T SIDUR: A Structuring Formalism for Knowledge Information Processing Systems %D 1984 %R CSTR-84-40-2 %I Oregon State University, Department of Computer Science %A S. Rehfuss %A M. J. Freiling %A J. Alexander %T Particularity in Engineering Data %D 1984 %R CSTR-84-40-3 %I Oregon State University, Department of Computer Science %X discusses inferential data in three engineering expert system databases %A Soveig A. Viste %A Chul E. Kim %T The Recognition of Digital Cylindrical Surfaces and Digital Moebius Strips %R CS-85-133 %I Washington State University Deparment of Computer Science %X $3.50 %A Y. Gurevich %A S. Shelah %T Fixed-Point Extension of First-Order Logic %R CRL-TR-5-85 %I Computing Research Laboratory, University of Michigan %D May 1985 ------------------------------ Date: 22 Nov 85 1432 PST From: Vladimir Lifschitz Subject: Seminar - The Riddle of STRIPS (SU) The Riddle of STRIPS and Its Solution Vladimir Lifschitz Non-Monotonic Reasoning Seminar Wednesday, November 27, 2pm MJH 252 STRIPS (STanford Research Institute Planning System) operates with world models represented by sets of formulas of first-order logic. A STRIPS system describes the effect of an action by a rule which defines how the current world model should be changed when the action is performed. The big mystery about STRIPS is why it does not produce incorrect results. Presumably, this happens because the rules correctly describe properties of the corresponding actions. But what do we mean by the "correctness" of a STRIPS rule? Straightforward attempts to define the semantics of STRIPS rules turn out to be unsatisfactory; we will examine a classical STRIPS system and show that, from some points of view, its rules are incorrect. The purpose of that exercise is not to criticize the system, but rather to show that defining the semantics of STRIPS is a tricky business. In the last part of the talk, a solution to the problem will be proposed. We will see that, under some conditions, STRIPS rules can be viewed as perfectly legitimate tools for formalizing knowledge about the effects of actions. ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Fri Dec 20 01:47:45 1985 Date: Fri, 20 Dec 85 01:47:30 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a000352; 3 Dec 85 17:09 EST Date: Tue 3 Dec 1985 13:02-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #181 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Wed, 4 Dec 85 05:00 EST AIList Digest Tuesday, 3 Dec 1985 Volume 3 : Issue 181 Today's Topics: Seminars - Truth Maintenance and Multiple Worlds in KEE (SU) & Model and Temporal Proof System for Processes (CMU) & Reasoning about Control in Vision (SRI) & An Approach to Conscious Experience (UCB), Conferences - Expert Systems and Their Applications & Knowledge and Data & Workshop on AI for Design Automation ---------------------------------------------------------------------- Date: Mon 2 Dec 85 08:54:31-PST From: Anne Richardson Subject: Seminar - Truth Maintenance and Multiple Worlds in KEE (SU) DAY December 3, 1985 EVENT Computer Science Colloquium PLACE Skilling Auditorium TIME 4:15 TITLE "Truth Maintenance and Multiple Worlds in KEE" PERSON Paul Morris, Robert Nado, Richard Fikes FROM IntelliCorp TRUTH, MAINTENANCE AND MULTIPLE WORLDS IN KEE We describe the integration of an assumption-based truth maintenance system (ATMS) into the frame-based representation facilities of the KEE system, and the use of the ATMS to implement a multiple-world context graph system for KEE. Integration into the frame system involves associating with potential slot values ATMS nodes that are used to determine in which worlds (contexts) the slot values are believed. Built-in inferences provided by the frame system, such as inheritance and the checking of value class and cardinality constraints, are recorded, when needed, as explicit justifications in the ATMS. In addition, the default reasoning capabilities of KEE have been refined and extended to take advantage of the ATMS. Tradeoffs in the integration between flexibility of use and run-time efficiency are examined. We describe the multiple-world context graph system with particular attention to an interpretation of the graph as a network of actions. In this framework, the semantics of graph merges are investigated and restrictions to ensure valid action sequences are discussed. ------------------------------ Date: 2 December 1985 1654-EST From: Theona Stefanis@A.CS.CMU.EDU Subject: Seminar - Model and Temporal Proof System for Processes (CMU) Date: Monday, 9 December Place: 5409 WeH Time: 3:30 PS SEMINAR A model and temporal proof system for processes Van Nguyen IBM Thomas J. Watson Research Center (joint work with Alan Demers, David Gries and Susan Owicki) There exist several models of processes, e.g. those of Brock-Ackerman, Hoare, Milner and Pratt. None of these models handles both synchronous and asynchronous communication in a single framework. In addition, their modeling of temporal properties (e.g. liveness properties) is generally unsatisfactory. The models that seem most promising, due to their simplicity and ability to hide information, are those based on traces. A trace is a finite sequence of communication events, which can be thought of as an abstraction of a process state in which all irrelevant internal details are hidden. A number of proof systems for processes have also been proposed. The Hoare-like proof systems, e.g. those of Chen-Hoare, Levin-Gries and Misra-Chandy, are simple but lack expressive power and cannot deal with temporal properties. Temporal proof systems, e.g. those of Manna-Pnueli and Barringer-Kuiper-Pnueli, are expressive but more complicated. We present a model for processes that is based on the notion of behavior (a generalization of trace). The model can handle either synchronous or asynchronous communication, and can describe temporal properties. We also describe a sound and complete temporal proof system that is based on the model. Due to the modularity of the model, the proof system is compositional. Both the model and proof system are simple. Thus we show that temporal proof systems can be made ------------------------------ Date: Tue 3 Dec 85 11:44:55-PST From: LANSKY@SRI-AI.ARPA Subject: Seminar - Reasoning about Control in Vision (SRI) REASONING ABOUT CONTROL IN A HIGH-LEVEL COMPUTER VISION SYSTEM Leonard Wesley SRI International, AI Center 11:00 AM, MONDAY, December 9 SRI International, Building E, Room EJ228 (new conference room) If you built an expert system, how would you expect it to decide what to do next in complex situations? Typically there are several alternative actions it might take to reach some goal. In some cases, the best alternative is clear or the choices do not warrant extensive analysis. At times the consequences of pursuing some action justify expending the effort to obtain the necessary information to analyze the pros and cons of choosing a particular alternative. Most would agree that the information that is needed to reach any decision is, to some degree, uncertain, imprecise, and occasionally inaccurate (called "evidential" information). Clearly knowledge about the certainty, precision, and accuracy of information can be used to improve a system's ability to reason about (i.e., control) its actions. In this talk, we shall describe how this might be accomplished by an expert system in the domain of high-level computer vision. We shall explain why we view Shafer's theory of belief functions as being better suited than some other models as a theoretical foundation for representing evidential information and reasoning about control. Results from a large number of image interpretation experiments will be presented to demonstrate how a system's performance can be improved when Shafer's theory is soundly exploited. Finally, we shall briefly describe how our approach to control might be extended to an evidential-based framework for planning under uncertainty. ------------------------------ Date: Mon, 2 Dec 85 09:43:33 PST From: admin%cogsci@BERKELEY.EDU (Cognitive Science Program) Subject: Seminar - An Approach to Conscious Experience (UCB) BERKELEY COGNITIVE SCIENCE PROGRAM Fall 1985 Cognitive Science Seminar - IDS 237A Tuesday, December 3, 11:00 - 12:30 240 Bechtel Engineering Center Discussion: 12:30 - 1:30 in 200 Building T-4 ``An Approach to Conscious Experience'' Bernard J. Baars Langley Porter Neuropsychiatric Institute, U.C.S.F. Conscious experience has been widely viewed as a confusing and ill-defined issue, and most psychologists have avoided it until quite recently. However, there are straightforward ways to specify reliable empirical constraints on the problem, sim- ply by contrasting comparable pairs of events, one of which is conscious and the other not. For example, we are typically unconscious of highly predictable stimuli, though there is strong evidence that such stimuli continue to be represented in the nervous system. We are unconscious of automatized actions, of the unattended stream in a selective attention paradigm, of conceptual presuppositions, of the unconscious meaning of per- ceptual and linguistic ambiguities, of lexical access, syntac- tic rule-application, etc. In all these cases the unconscious information continues to be represented and processed. Any complete theory of conscious experience is bounded by, and must ultimately account for, the entire set of such contrasts. The empirical constraints converge on a model of the ner- vous system as a distributed collection of specialists--- automatic, unconscious, and very efficient. Consciousness is associated in this system with a "global workspace"---a memory whose contents are broadcast to all the specialists. Special- ists can complete or cooperate for access to the global workspace, and those that succeed can recruit and control other specialists in pursuit of their goals. Over the past seven years this Global Workspace approach has been extended to a number of puzzling issues, including action control and the neurophysiological basis of consciousness. ------------------------------ Date: Saturday, 30 November 1985 21:06:58 EST From: Duvvuru.Sriram@cive.ri.cmu.edu Subject: Conference - Expert Systems and Their Applications SIXTH INTERNATIONAL WORKSHOP ON EXPERT SYSTEMS & THEIR APPLICATIONS INTRODUCTION Following the success of the 5th International Workshop in Expert Systems and their Applications at the prestigious 14th century fortess Palace of the Popes in Avigon (France), the Agence de l'Informatique has scheduled the 6th Workshop for April 28-30, 1986. Papers are solicited which describe expert systems actually applied in industry, currently under assessment by users, or currently commercially available. INSTRUCTION TO AUTHORS Five copies of submitted papers (not exceeding 20 pages in 8 x 11" camera-ready format) should reach the Workshop Chairman before January 15, 1986. Papers may be written either in English or French. Simultaneous translation will be provided at the conference. Submitted papers should include a page with: - title of the paper - author's name - author's address - phone number and extension - telex number - a 10 line abstract - a list of key words All papers will be refereed by an international program committee; notification of acceptance will be given by March 4, 1986. A best paper award will be presented at the conference. OTHER ACTIVITIES Tutorials and panel discussions are planned. Send suggestions for topics to Workshop Chairman at the address indicated. INFORMATION For further information contact: Jean-Claude Rault Workshop Chairman Agence de l'Informatique Tour Fiat - Cedex 16 92084 Paris - La Defence France Tel. (331) 47 96 43 14 ------------------------------ Date: Sat, 30 Nov 85 19:05:15 EST From: "John F. Sowa" Subject: Conference - Knowledge and Data IFIP INTERNATIONAL FEDERATION FOR INFORMATION PROCESSING ANNOUNCEMENT TC2 WORKING CONFERENCE organized by Working Group 2.6 Knowledge and Data (DS-2) November 3-7, 1986 in Albufeira (Algarve), Portugal Scope: Questions of meaning are more important for the design of a knowledge base than methods of encoding data in bits and bytes. As database designers add more semantic information to their systems, their conceptual schemata begin to look like AI systems of knowledge representation. In recognizing this convergence on issues of semantics, IFIP Working Group 2.6 is organizing a working conference on Knowledge and Data. It will address the issues and problems of knowledge representation from an interdisciplinary point of view. Topics: Design of a conceptual schema Knowledge and data modeling Database semantics Natural language semantics Expert database systems Logic, databases, and AI Methods of knowledge engineering Tools and aids for knowledge acquisition Invited speakers: Herve Gallaire, Germany Robert Meersman, Belgium J. Alan Robinson, USA Roger Schank, USA Dana Scott, USA An IFIP working conference is oriented towards detailed discussion of the topics presented. Participation is by invitation, with optional contribution of a paper that is refereed by the program committee. Anyone who is interested in participating should send an abstract of current research or a prospective paper to either of the program cochairmen. Abstracts are due March 14, 1986. Complete papers are due May 16, 1986. Papers presented at the conference will be published in book form by North Holland Publishing Co. General Chairman: Amilcar Sernadas, Portugal Program cochairmen: John F. Sowa Robert Meersman IBM Systems Research Institute L.U.C. -- Dept. WNIF 500 Columbus Avenue Universitaire Campus Thornwood, NY 10594 B-3610 Diepenbeek U.S.A. Belgium CSNET: sowa.yktvmt@ibm ------------------------------ Date: Tue, 26 Nov 85 15:58:44 est From: Scott C McKay Subject: Conference - Workshop on AI for Design Automation FIRST AFWAL* RESEARCH WORKSHOP TO DEVELOP AND AUTOMATE A SCIENCE OF DESIGN FOR MILITARY WEAPONS SYSTEMS VIA APPLIED ARTIFICIAL INTELLIGENCE Georgia Tech Research Institute Atlanta, Georgia March 24-26, 1986 The premise of this workshop is that appropriate research can create a new, invented Science of Design to support CAE/CAD in a CAE/CAD/CAM Military Weapons System Foundry. A Foundry is required for rapid design and production of complex weapons systems demanded by changing military mission requirements. CAE/CAD is viewed as a domain of Applied AI called Design Automation (DA), and Design Science is considered a subdiscipline of Design Automation. DA is viewed as a totally generic discipline whose domain is both weapons systems and their embedded electronics, including multiplatform systems. The discipline of DA incorporates requirements engineering, weapons system and subsystem configuration, the design of mission-specific system and subsystem functions, signal and data processing algorithm design, software engineering (including firmware), multiprocessor and processor design, and structural, mechanical, thermal, electrical, electromagnetic, and electronics (including analog) engineering. For this workshop the domain of DA will be limited to military embedded electronics systems (including multiplatform systems). Within this domain the workshop is generic, in that it includes all the preceding DA disciplines from requirements engineering to electronics engineering. The workshop's purpose is to prepare approximately five detailed near-term project plans for initial vectoring of DA-relevant research towards Weapons System Foundry objectives. One project plan will detail development of a DA Testbed. Additional special interest project plans may also be prepared. All workshop attendance expenses are the responsibility of attendees, and include a nominal registration fee for lunches and refreshments. Only US citizens will be allowed to attend the workshop. There is no a priori restriction on attendees' technical background or employer. An attendance limit of 40 participants may cause rejection of some attendance applications; however, all applicants will be provided copies of the draft workshop report for review. An Attendance Application Package is available on request. It contains a white paper defining AFWAL objectives for long term DA technology development, a detailed workshop description, and a questionnaire to establish individual applicant's DA-relevant qualifications and interests. To request an Attendance Application Package please contact: Mr Harold Noffke AFWAL/AARM-3 WPAFB OH 45433-6543 Tel: (513) 255-3655/5097/6071 *AFWAL stands for Air Force Wright Aeronautical Laboratories. ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Wed Jan 1 11:52:46 1986 Date: Wed, 1 Jan 86 11:52:25 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a004805; 5 Dec 85 3:26 EST Date: Wed 4 Dec 1985 23:30-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #182 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Thu, 5 Dec 85 07:05 EST AIList Digest Thursday, 5 Dec 1985 Volume 3 : Issue 182 Today's Topics: Queries - Knowledge Base Management System & Machine Learning & Subconscious Reasoning & Decision Theory & Natural Language & Distributed Computing using AppleTalk, AI Tools - Smalltalk & Object-Oriented Programming in Lisp, Expert Systems - Artificial Empathy, Programming Languages - Type Checking ---------------------------------------------------------------------- Date: Tue, 3 Dec 85 22:10:06 est From: mex107@mitre (Michael Leavitt) Subject: First use of phrase "knowledge base management system" I'm trying to track down the first use of the phrase "knowledge base management system" to refer to expert system shells like KEE and ART. I've heard it in oral presentations, but not seen it in writing. Can anyone help? Replies to me please. Many thanks. Mike Leavitt ------------------------------ Date: Fri, 29 Nov 85 12:48 CDT From: Joseph_Tatem Subject: Query - paper on machine learning In Artificial Intelligence #25, Kurt VanLehn, reviewing Machine Learning, says "Carbonell's more recent papers report that further implementation uncovered fundamental flaws in the design presented [in Carbonell's chapter in Machine Learning]". Can anyone direct me to these papers and/or give me an idea what theses flaws are?? Thanx, Joe Tatem tatem.ti-eg@csnet-relay ------------------------------ Date: Fri, 29 Nov 85 11:46:44 GMT From: gcj%qmc-ori.uucp@ucl-cs.arpa Subject: Subconscious Reasoning : Discovery and Invention subconscious, adj. ``of our own mental activities of which we are not aware'', n. ``the part of the mind in which these activities take place.'' The Oxford Paperback Dictionary. What does cognitive science have to make of the discovery of the structure of the benzene molecule? It is said to have been the result of a chemist having a dream in which serpents were biting their own tails. He interpreted this when waking as the long sought after ring configuration that would make sense of the ratios of hydrogen to carbon atoms that had been discovered by experiment. Gordon Joly gcj%qmc-ori@ucl-cs.arpa [An even more impressive feat was the translation of Samuel Pepys' diary, written in a code (actually an ancient shorthand, for which the key was later found in his personal library) which had resisted many cryptographic efforts. The owner of the diary one night dreamt a full page of cleartext, including the page number, and was then able to translate the whole thing. -- KIL] ------------------------------ Date: 27-NOV-85 From: PETER PIRRON Subject: Request - Decision Theory and Natural Language HALLO I am a new member in the AIList. My name is Peter Pirron. I am living in Heidelberg/West Germany. I am working as an computer advicer for statistical problems at the Psychochological Institut. At the moment my interests are normative decision theory and programming of natural language. It would be very helpful for me if you could send me references of articles and books of these subjects. It would also be interesting for me to know people who work on these subjects. I am especially interested in programs dealing with the subjects mentioned above. Thank you very much in advance!!! Peter Pirron (h29@dhdurz2.bitnet) ------------------------------ Date: Wed, 4 Dec 85 13:25 PST From: "Watson Mark%SAI.MFENET"@LLL-MFE.ARPA Subject: SMALLTALK, LISP, DISTRIBUTED COMPUTING, APPLETALK, MACINTOSH Date: 12/4/85 FROM: WATSON#M%SAI@LLL-MFE Subject: Lisp and Smalltalk on the Macintosh In response to Fabio Favata's request for a public domain version of the Smalltalk language, I would suggest contacting Apple Computer Inc. Apple has made the original Xerox Smalltalk language available on the Macintosh for $50 (for 7 disks!!). The Smalltalk system runs on a 512K Mac. To order Smalltalk, call 1 408 747-1288 and ask for Paula. I would like to hear from anyone attempting to use the Appletalk network to build testbed distributed computing systems. The Appletalk network has a fairly low bandwidth (.25 megabits/sec), but the availability of two good symbolic programming languages on the Mac (Smalltalk and ExperLisp) would provide a very low cost system for software development. I can be reached at 1 619 456-6816. - Mark Watson ------------------------------ Date: Wednesday, December 4, 1985 17:11:12 From: snyder@hplabsd Subject: Object-Oriented Programming in Lisp From: Nick Davies (at GEC Research) Subject: Object oriented programming in Common Lisp > Does anyone have or know of an implementation of Flavors or any other > object-oriented programming system in Common Lisp ? Hewlett-Packard has developed an object-oriented extension to Common Lisp, which we are proposing as a candidate for standardization. Our extension is similar in syntax to Flavors, but is stricter in its support of encapsulation. Like Common Lisp itself, our extension has been designed to serve as a common language subset for object-oriented programming (in Common Lisp). It also has been designed to permit a very efficient implementation, even on conventional hardware. At Hewlett-Packard, we have been using object-oriented programming in Lisp for many years. The Common Lisp version of our objects dialect has been in use since last February. Our AI workstation software, which was demonstrated at IJCAI in August, uses object-oriented programming heavily in its extensible user interface, and in applications. For example, the configuration I am running contains 132 class definitions. The object-oriented extensions will be included in our forthcoming Common Lisp development product for HP Series 300 workstations (68020 based). To receive a copy of a memo describing our Common Lisp object-oriented extensions, send your U.S. mail address to Mingus%hplabs@csnet-relay.arpa and ask for ATC-85-1, "Object-Oriented Programming for Common Lisp". Alan Snyder ------------------------------ Date: Wed, 27 Nov 85 11:39:39 mst From: ted%nmsu.csnet@CSNET-RELAY.ARPA Subject: expert systems move on TI just distributed a marketing flyer in which they described expert systems AT WORK TODAY (original emphasis) solving problems like: ... credit approval ... policy and procedure administration It seems that the marketing folks at TI haven't heard about the discussions recently on the digest about how artificial empathy will be required to properly implement these sorts of applications. ------------------------------ Date: Wed, 27 Nov 85 18:23:01 CST From: reddy@a.CS.UIUC.EDU (Uday S. Reddy) Subject: Type checking I do not see why it would be easier to locate "type errors" (errors detectable by a type checker) than other kinds of errors. The effects of errors get propagated the same way and their symptoms are exhibited the same way. My experience with programming in LISP was that I was indeed spending an inordinate amount of time locating errors that could have been detected by a type checker. The real advantage of a type checker is that it not only detects type errors, but also locates them. While detection can be done by run time type checking, location is not done. ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Wed Jan 1 12:34:05 1986 Date: Wed, 1 Jan 86 12:33:52 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a028036; 10 Dec 85 2:38 EST Date: Mon 9 Dec 1985 22:24-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #183 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Tue, 10 Dec 85 21:37 EST AIList Digest Tuesday, 10 Dec 1985 Volume 3 : Issue 183 Today's Topics: Seminars - CAKE Knowledge Representation and Reasoning System (MIT) & - SOJA Scheduling System (CMU) & - Fault Analysis using Dynamic Behavior (Rutgers) & - Toward a Theory of Algorithm Improvement (Oregon State) & - Inheritance, Data Models and Data Types (MIT) & - STRIPS and Circumscription (SU) & - Semi-Applicative Programming (SRI) & - Connectionist Talk Talk (SU) & - Possible Worlds and Situations Semantics (UPenn) ---------------------------------------------------------------------- Date: Mon, 2 Dec 85 09:22 EST From: Brian C. Williams Subject: Seminar - CAKE Knowledge Representation and Reasoning System (MIT) [Forwarded from the MIT bboard by SASW@MIT-MC.] Thursday , December 5 4:00pm Room: NE43- 8th floor Playroom The Artificial Intelligence Lab Revolving Seminar Series CAKE: An Example of a Hybrid Knowledge Representation and Reasoning System Charles Rich MIT AI Lab Cake is a knowledge representation and reasoning system being developed to support automated programming (the Programmer's Apprentice). The first part of this talk describes the architecture of Cake, which is divided into the following nine layers, each with associated representations and reasoning procedures: Plan Synthesis Plan Recognition Plan Calculus Frames Types Algebraic Demons Equality Truth Maintenance The second part of the talk takes a look at some of the issues in the design of hybrid systems generally, such as What is a hybrid system? Why would you want one? Who is developing them? Where do we go from here? Nelson & Oppen's cooperating decision procedures, KL-Two, Krypton, and Cake will be discussed as examples of hybrid systems. ------------------------------ Date: 4 Dec 85 11:30:50 EST From: Jeanne.Bennardo@ISL1.RI.CMU.EDU Subject: Seminar - SOJA Scheduling System (CMU) Intelligent Systems Lab Seminar Topic: Presentation of SOJA project. Speaker: Claude Le Pape Place: DH3313 Date: Friday, Dec. 6 Time: 10:30am - 11:30am Speaker: Claude Le Pape is a visiting researcher from the Laboratoires de Marcoussis in France, and is currently working in the Intelligent Systems Lab with the Phoenix/Opis project. Abstract: We have implemented in the Laboratoires de Marcoussis - in Franz Lisp - a daily scheduling system called SOJA. Given the state of the shop in the evening, SOJA builds a scheduling plan for the next day. This implies : - selecting the operations to be performed over the day - scheduling them and computing a time-table for each machine. Two planning steps are distinguished in SOJA : Selection and Scheduling. - Selection is done according to the state of the shop and the orders to be completed. Selection rules are used to build a tagged graph, which nodes are phases (operations). Each arc is tagged according to a reason for selecting its extremity, and valued in order to measure the significance of this reason. Then SOJA examines this graph and the resource requirements to decide which phases should be scheduled. - Scheduling is considered as a constraint-directed reasoning task. However, preferences are not represented as constraints, but as scheduling rules that are used to make decisions. Both selection and scheduling processes use an inference engine that has been specially designed for SOJA. We will describe this engine and the solutions we have implemented to make it efficient : - combination of forward and backward chaining - preservation of partial instances of rules (and propagation) - compilation of rules. We will also discuss our constraint representation and backtracking. A short comparison between SOJA and ISIS is also scheduled. ------------------------------ Date: 4 Dec 85 10:38:59 EST From: Smadar Subject: Seminar - Fault Analysis using Dynamic Behavior (Rutgers) III Seminar Title: Using Dynamic Behavior of Physical Systems for Fault Diagnosis Speaker: Kathy Abbott Date: Thursday, December 5, 1985, 11:30-12:30PM (*** notice date and time change ***) Place: Hill Center, room 423 Kathy Abbott, a Ph.D. candidate in our department and researcher at NASA - Langley, will discuss her dissertation research on fault diagnosis. Here is her abstract: One consideration when performing real-time diagnosis of physical systems is that the system's behavior may change as time progresses. The effect of a failure may propagate through the system under consideration as well as to systems either functionally or physically adjacent to it. This dynamic behavior can be used to prune fault hypotheses by using models of the physical system to identify how the fault is propagating, the type of fault, and what components of the physical system are affected. An advantage of using functional and physical models in this manner is that it permits reasoning about incomplete information, such as, lack of measurements due to physical damage, system parameters that are not sensed, or sampling times that are not frequent enough to detect all system changes. In this talk I will discuss a model-based diagnosis system which uses the dynamic behavior of physical systems in the diagnosis process. The diagnosis system is not yet completely fleshed out in all details, so any constructive comments or suggestions are particularly welcome at this time. ------------------------------ Date: Thu, 5 Dec 85 14:00:35 pst From: Tom Dietterich Subject: Seminar - Toward a Theory of Algorithm Improvement (Oregon State) This colloquium has already happened, but I thought you might still be interested in sending it to AILIST. --Tom Departmental Colloquium Oregon State University Department of Computer Science Wednesday, Nov. 20, 1985 TOWARD A THEORY OF ALGORITHM IMPROVEMENT Thomas G. Dietterich Department of Computer Science Oregon State University This talk will sketch a theory of algorithm improvement called the "test incorporation" theory. The basic premise of test incorporation theory is that all algorithms can be usefully viewed as improvements of a naive generate-and-test algorithm. The improvements can take one of two forms: (a) movement along the space-versus-time tradeoff curve and (b) INCORPORATION of test information into the generator of possible solutions. Examples will be presented for each kind of performance improvement. The talk will then take up the questions of WHEN these improvements take place and WHAT PROGRAM is performing the improvements. Traditional compilers perform simple kinds of improvements at "compile-time". However, many more powerful algorithms make these improvements to themselves dynamically at run time, particularly when new input data is made available. Test incorporation theory can be employed in two ways: (a) to analyze existing programs and understand why they are successful, and (b) to design new programs. The test incorporation perspective also provides some interesting answers to the questions "What is computer science?" "What is knowledge?" and "What is intelligence?" This work is a joint project with James Bennett of Teknowledge, Inc. ------------------------------ Date: Thu, 5 Dec 85 16:21:29 est From: nikhil at MIT-NEWTOWNE-VARIETY.MIT.EDU (Rishiyur S. Nikhil) Subject: Seminar - Inheritance, Data Models and Data Types (MIT) [Forwarded from the MIT bboard by SASW@MIT-MC.] Inheritance, Data Models and Data Types Peter Buneman University of Pennsylvania The notion of type inheritance (subsumption, ISA hierarchies) has long been recognised as central to the development of programming languages, databases and semantic networks. Recent work on the semantics of programming languages has shown that inheritance can be cleanly combined with functional programming and can itself serve as a model for computation. Using a definition of partial functions that are well behaved with respect to inheritance, I have been investigating a new characterization of the relational and functional data models. In particular, I want to show the connections of relational database theory with type inheritance and show how both the relational and functional data models may be better integrated with typed programming languages. Host: Prof. Rishiyur Nikhil Date: Friday, December 6, 1985 Time: 2:00 pm - Refreshments 2:15 pm - Lecture Place: Room NE43-512A MIT Laboratory for Computer Science, 545 Technology Square, Cambridge, MA 02139 ------------------------------ Date: 05 Dec 85 1134 PST From: Vladimir Lifschitz Subject: Seminar - STRIPS and Circumscription (SU) STRIPS and Circumscription: Two Approaches to the Frame Problem Vladimir Lifschitz Wednesday, December 11, 2:00 MJH 252 The frame problem consists in defining which properties of situations do not change across events. We will compare two well-known attempts to solve the frame problem. One, the STRIPS approach, is based on using a language which has no explicit references to situations. The other approach uses circumscription to express the "commonsense law of inertia", according to which the differences between two situations separated by an event are minimal, given the properties of the event. We will show on a simple example how to transform a circumscriptive theory into the description of the corresponding STRIPS operator. ------------------------------ Date: Fri 6 Dec 85 15:05:51-PST From: Phil Cohen Subject: Seminar - Semi-Applicative Programming (SRI) N. S. Sridharan, of BBN Labs, will give a seminar next Tues, the 10th, 10:30 am, in EJ228. Title and abstract below. Title: Semi-applicative Programming Most current parallel programming languages are designed with a sequential programming language as the base language and have added constructs that allow parallel execution. We are experimenting with an applicative base language that has implicit parallelism everywhere, and then we introduce constructs that inhibit parallelism. The base language uses pure LISP as a foundation and blends in interesting features of Prolog and FP. Proper utilization of available machine resources is a crucial concern of programmers. We advocate several techniques of controlling the behavior of functional programs without changing their meaning or functionality: program annotation with constructs that have benign side-effects, program transformation and adaptive scheduling. This combination yields us a SEMI-APPLICATIVE programming language and an interesting programming methodology. Starting with the specification of a context-free recognizer, we have been successful in deriving variants of the recognition algorithm of Cocke-Kasami-Younger. One version is the CKY algorithm in parallel. The second version includes a top-down predictor to limit the work done by the bottom-up recognizer. The third version uses a cost measure over derivations and produces minimal cost parses using a dynamic programming technique. In another line of development, we arrive at a parallel version of the Earley algorithm. ------------------------------ Date: Fri, 6 Dec 85 06:03:17 pst From: gluck@SU-PSYCH (Mark Gluck) Subject: Seminar - Connectionist Talk Talk (SU) NETTALK: Teaching a Massively-Parallel Network to Talk Dec. 18th, 1:00 pm, at Redwood hall Terrence J. Sejnowski Biophysics Department Johns Hopkins University Baltimore, MD 21218 ABSTRACT Text to speech is a difficult problem for rule-based systems because English pronunciation is highly context dependent and there are many exceptions to phonological rules. An alternative knowledge representation for correspondences between letters and phonemes will be described in which rules and exceptions are treated uniformly and can be determined with a learning algorithm in a connectionist model. The architecture is a layered network of 400 simple processing units with 9,000 weights on the connections between the units. The training corpus is continuous informal speech transcribed from tape recordings. Following training on 1000 words from this corpus the network can generalize to novel text. Even though this network was not designed to mimic human learning, the development of the network in some respects resembles the early stages in human language acquisition. Following damage of the network by either removal of units or addition of random values to the weights the performance of the network degraded gracefully. Issues which will be addressed include scaling of the learning algorithm with the size of the problem, robustness of learning to predicate order of the problem, and universality of learning in connectionist models. ------------------------------ Date: Mon, 9 Dec 85 11:58 EST From: Tim Finin Subject: Seminar - Possible Worlds and Situations Semantics (UPenn) Forwarded From: Dale Miller on Mon 9 Dec 1985 at 8:58 Joint Mathematics / Computer Science LOGIC COLLOQUIUM Possible Worlds and Situations Semantics Greg Hager CIS Department, UPenn Monday 9 December 1985 4:40 pm, DRL 4E17 Situation theory is an attempt by its developers (philosopher John Perry and mathematician Jon Barwise) to provide an alternative framework for the study of language and meaning. Their approach has been to start over from first principles and construct a theory which does not rely on classical logic as developed by Frege -- in particular emphasizing the notions of partial information, a relational theory of meaning, and mathematical realism. Needless to say this has provoked substantial interest, debate, and skepticism from a variety of sources. Any complete understanding of pros and cons of this theory requires a substantial knowledge of philosophy, mathematics, and linguistics (none of which I can lay claim too). However, from one point of view, situation theory can be loosely construed as an alternative to possible worlds theory and its application to the study of language by Montague. In this talk, I will focus on the development of possible worlds semantics and its use in linguistics, and point out where situation theory diverges or disagrees. This discussion will be fairly general and open-ended, so audience participation is welcomed and encouraged. ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Wed Jan 1 12:44:41 1986 Date: Wed, 1 Jan 86 12:44:32 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a028829; 10 Dec 85 4:36 EST Date: Mon 9 Dec 1985 22:31-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #184 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Tue, 10 Dec 85 06:06 EST AIList Digest Tuesday, 10 Dec 1985 Volume 3 : Issue 184 Today's Topics: Query - Translator: SAIL --> LISP?, AI Tools - Object-Oriented Programming in Common Lisp, Bindings - AI&DS Name Change, Journal - Machine Learning ---------------------------------------------------------------------- Date: Sun 8 Dec 85 00:59:07-PST From: Kashi.Rao%dworkin@usc-eclc.ARPA, 220P%dworkin@usc-eclc.ARPA, Subject: Translator: SAIL --> LISP? I have a large program in SAIL that I would like to convert to LISP. I would like to know if there's any program available to translate code from SAIL to LISP (COMMON or FRANZ). Any quick pointers/references/suggestions would be deeply appreciated. Kashi Rao Intelligent Systems Group, USC. rao%dworkin@usc-oberon ------------------------------ Date: Thu, 5 Dec 85 14:52 CDT From: Joseph_Tatem Subject: RE: Object-Oriented Programming in Common Lisp The TI Explorer supports Flavors in Common Lisp. Joe Tatem ------------------------------ Date: Thu, 5 Dec 85 13:33:35 pst From: clif@aids-unix (Clif McCormick) Subject: Name Change Advanced Decision Systems 201 San Antonio Circle Suite 286 Mountain View, California 94040 (415) 941-3912 Advanced Information & Decision Systems, a leader in defense and commercial applications of artificial intelligence and other advanced information processing technologies, has changed its name to: ADVANCED DECISION SYSTEMS Founded in 1979, Advanced Decision Systems specializes in applied research and prototype development involving a number of technologies, including artificial intelligence, computer science, estimation and control theory, signal and image processing, and decision theory. Specific projects include image understanding for scene analysis and object classification, oil exploration, fault diagnosis, distributed sensor networks, multi-target tracking, autonomous land vehicle navigation, avionics expert systems, and a battlefield commander's assistant. Primary among the clients for these projects are defense organizations such as the Department of Defense's Advanced Research Projects Agency, and research laboratories of the U.S. Army, Navy, and Air Force. The name change comes as Advanced Decision Systems seeks to expand its involvement in the commercial marketplace and add to its list of commerical clients, which currently includes Bank of America, McGraw-Hill, and Kodak. Advanced Decision Systems, with headquarters in Mountain View, California and a recently opened office in Arlington, Virginia, is completely employee-owned and had revenues of $6.5 million in fiscal 1985. The company currently has a total of 86 employees. The name change, from Advanced Information & Decision Systems to Advanced Decision Systems, will become effective December 1, 1985. Thanks. Clif McCormick clif@aids-unix (415) 941-3912 ------------------------------ Date: 5 December 1985 0703-PST (Thursday) From: west@nprdc.arpa (Larry West) Subject: New Journal: ``Machine Learning'' I'm sure many of you received the same flyer I did, but for those who didn't, the new journal ``Machine Learning'' is about to begin its first volume, and is soliciting subscriptions and submissions. Executive Editor: Pat Langley, UCI Editors: Jaime G. Carbonell, CMU Ryszard S. Michalski, UIllinois Tom M. Mitchell, Rutgers The Editorial Board: Saul Amarel, John R. Anderson, Bruce G. Buchanan, Thomas G. Dieterrich, Geoff Hinton, John H. Holland, Yves Kodratoff, Douglas B. Lenat, Jack Mostow, J.R. Quinlan, Paul S. Rosenbloom, Roger Schank, Derek Sleeman, and the omnipresent Patrick Winston. Quoting: Machine Learning will publish papers on the processes through which intelligent systems improve their performance over time, and will cover areas such as: * Concept Acquisition * Strategy Learning * Language Development * Reasoning by Analogy * Scientific Discovery The main emphasis will be on symbolic learning methods as opposed to numeric ones [boo], and the editors will give preference to papers that describe theories or principles of learning that are supported by running computer programs. However, the journal also welcomes theoretical treatments and comparisons of existing systems, and encourages computational models of human learning. Although the journal will emphasize basic research in the emerging field of Machine Learning, it will also consider applications of learning methods to real-world domains. ... Machine Learning publishes articles on the processes through which intelligent systems improve their performance over time. Authors are invited to submit papers describing computational approaches to any aspect of learning. We encourage attempts to formulate principles of learning to integrate previous research results, and to model human learning in a computational framework. Four issues per year, $35 individual, $78 institutional; first issue due out in January 1986. Kluwer Academic Publishers 190 Old Derby Street Hingham, MA 02043 617-749-5262 (for Credit Card orders) Larry West USA+619-452-6771 Institute for Cognitive Science non-business hrs: 452-2256 UC San Diego (mailcode C-015) La Jolla, CA 92093 USA UUCP: {ucbvax,ihnp4,sdcrdcf,decvax,gatech}!sdcsvax!sdcsla!west or {sun,ulysses}!sdcsla!west ARPA: or DOMAIN: or ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Wed Jan 1 12:32:52 1986 Date: Wed, 1 Jan 86 12:32:34 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a000034; 13 Dec 85 1:34 EST Date: Thu 12 Dec 1985 22:16-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #185 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Fri, 13 Dec 85 22:57 EST AIList Digest Friday, 13 Dec 1985 Volume 3 : Issue 185 Today's Topics: Queries - Expert Systems for Building Configuration & LIPS, Policy - AI&DS Name Change and Advertising, Literature - Spang Robinson Report No. 2 & Proceedings of the 1st Annual Conference on Law and Technology, Seminars - Parallel Depth-First Search (MIT) & Quantitative Operators in Design (UTexas) & An Object Model of Information (CMU) & Massively Parallel Networks that Learn Representations (MIT) ---------------------------------------------------------------------- Date: 11 December 1985 1207-EST From: William Proffer@A.CS.CMU.EDU Subject: Bldg Config. KBES I'd like to know if anyone has any information on expert systems used in the configuration of office buildings for tenants. For instance, configuration of utilities, space, parking, etc. Please mail replies to proffer@s3sun.css.gov Thanks. Bill Proffer ------------------------------ Date: Tue, 10 Dec 85 21:08 CST From: Jerry Bakin Subject: fLIPS, pLIPS, nLIPS, uLIPS, mLIPS, LIPS, KLIPS, MLIPS, GLIPS, In many Prolog discusions, I have heard people talk about the speed of an implementation in terms of Logical Inferences per Second (LIPS). I don't recall ever seeing a formal definition of this. Would someone care to discuss what a LIP is? Is it one unit of unification? Some sort of average time it takes prolog to scan a clause? As benchmarks go, is apples and oranges, or red delicious and golden delicious? How can it be compared to lisp? In terms of LIPS, how fast are current Prolog implementations? DEC-10? DEC-20? Symbolics? Micro-Prolog? And finally, in terms of LIPS, how fast are LISP implementations for the DEC-20 and Symbolics (or other machines)? Thanks, Jerry Bakin ------------------------------ Date: Tue, 10 Dec 1985 05:50 CST From: AI.DUFFY@R20.UTEXAS.EDU Subject: Name Change: Banned AIDS! Date: Thursday, 5 December 1985 15:33-CST From: clif at aids-unix (Clif McCormick) Advanced Information & Decision Systems ... advertising deleted ... has changed its name to: ADVANCED DECISION SYSTEMS ... advertising deleted ... The name change comes as Advanced Decision Systems seeks to expand its involvement in the commercial marketplace and add to its list of commerical clients, which currently includes ... assorted multinationals deleted ... Come on, tell us WHY you've changed your name! All you've done is removed the word INFORMATION from your name. What do you have against information? Are you trying to promote RANDOMNESS? Or maybe it's your commercial clients who don't like information. Is that it? Hmmmm.... Maybe its the acronym. Maybe you don't like the acronym AIDS because you're not really in the business of AIDing anybody (except of course your commercial clients) and you don't want people bothering you whenever they need some AID. You've now changed the acronym to ADS. So that must be it. You aren't against information at all. You've just decided to change your name to ADS to let people know you like to place ADS on AIList. Right? [The original message was sent to AIList as a "press release". I chose to run it under "Bindings", although much of it fell under "Lab Descriptions". I have permitted such reports from other sites (as does the SIGART Newsletter) -- I usually decide in favor of the transmission of new information and the suppression of repeated information, but do screen out job ads and anything that I consider blatantly commercial. My criteria have been partly shaped by the feedback or lack thereof from readers, and such feedback is always welcome. -- KIL] ------------------------------ Date: 10 Dec 1985 18:03-CST From: leff%smu.csnet@CSNET-RELAY.ARPA Subject: Spang Robinson Report, No 2 Summary of Spang Robinson Report, Dec 1985 Vol 1 NO. 2 AI AND MANUFACTURING Carnegie makes 70 percent of its 8 million in revenue from custom applications to clients in manufacturing. Teknowledge makes 50 percent of its revenues from custom consulting 30 percent of KEE applications are in manufacturing. A. D. Little has developed a "Smart Foreman" expert system Arthur Anderson has developed KIOSK, which helps process special orders, and Cell Designer which assigns parts and assigns machines to processes and cells. Applicon has demonstrated a knowledge based CAD/CAM workstation for sheet metal fabrication. International Systems Operations is developing systems to assist in quality control and material movement. Cognition is developing a system that captures design knowledge in handbooks and standard textbooks as well as corporate design standards. TIPNIS will provide manufacturing expertise to the company's designers. LMI will offer connectivity between PICON, it's real-time expert system tool and MAP, the new standard for factory floor networking. This will involve a < $7,000 board set. REVIEW of Texas Instruments satellite AI broadcast. News Section: ICONICS has changed its name to Transform LOGIc corporation. They are devleoping an AI-based software engineering system. RCA has established an AI chair at the University of Pennsylvania. VOTAN has announced a voice system for entering or verifying data which costs $4950.00. Carnegie Group has introduced Version 3 of Knowledge Craft. Artelligence and Computer*Thought have filed a law suit regarding breaches of non-competition agreements. Teknowledge has named Michael P. Chan, vice president of product sales for its Knowledge Engineering LISP machine has elected three new members of its board of directors. Summary of Artificial Intelligence: Towards Practical Applications ------------------------------ Date: Tue, 10 Dec 85 15:09:29 pst From: George Cross Subject: Proceedings: 1st Annual Conference on Law and Technology Here is the table of contents of the recently published proceedings of the First Annual Conference on Law and Technology, held at the University of Houston in August 1984 and sponsored by the University of Houston Law Center Program on Law and Technology. There was a Second Conference in 1985, but the proceedings are not available yet. The book is: %B Computing Power and Legal Reasoning %E Charles Walter %I West Publishing Company %C St. Paul, MN %D 1985 ISBN 0-314-95570-4 and is available from West Publishing Company 50 West Kellogg Boulevard P.O. Box 64526 St. Paul, MN 55164-1003 - - - - - - - - - - - - - - - - - - - - - - %A Layman Allen %A Charles Saxon %T Computer-Aided Normalizing and Unpacking: Some Interesting Machine-Processable Transformations of Legal Rules %P 495-572 %A Kevin Ashley %T Reasoning by Analogy: A Survey of Selected AI Research with Implications for Legal Expert Systems %P 105-127 %A Helene Bauer-Bernet %T Beyond Keyword Interaction: Computerized European Community Law %P 337-374 %A Daniele Bourcier %T About Intelligence in Legal Information Systems %P 319-336 %A William Boyd %T Choosing Between a Chapter 7 and a Chapter 13 Bankruptcy: An "Expert System" to Assist an Attorney in Making the Choice %P 699-763 %A Constantino Ciampi %A Deirdre Exell Pirro %A Elio Fameli %A Giuseppe Trivisonno %T THES/BID: An Expert System for Constructing a Computer-Based Thesaurus for Legal Informatics and Computer Law %P 375-412 %A Cary deBessonet %A George Cross %T Representation of Some Aspects of Legal Causality %P 205-214 %A Michael Dyer %A Margot Flowers %T Toward Automating Legal Expertise %P 49-68 %A Jerald Feinstein %T A Knowledge-Based Expert System Used to Prevent the Disclosure of Sensitive Information at the United States Environmental Protection Agency %P 661-697 %A Thomas Gordon %T Object-Oriented Predicate Logic and Its Role in Representing Legal Knowledge %P 163-203 %A Grayfred Gray %T Statutes Enacted in Normalized Form: The Legislative Experience in Tennessee %P 467-493 %A Grayfred Gray %T Law & Technology Conference Expert System Workshop Report %P 621-626 %A Michael Heather %T Demand-Driven Model for a Half-Intelligent System %P 69-103 %A Jay Hook %T Semantic Representations of Children's Blame Rules %P 29-47 %A Robert Krovetz %T The Use of Knowledge Representation Formalisms in the Modeling of Legal Concepts %P 275-317 %A Sydney Lamb %T Information and its Representation in English Texts %P 145-155 %A C. Duncan MacRae %T Tax Problem Solving with an If-Then System %P 595-620 %A Antonio Martino %T Why an Automated Analysis of Legislation? %P 413-466 %A L. Thorne McCarty %T Permissions and Obligations %P 573-594 %A Mark Peterson %A Donald Waterman %T An Expert Systems Approach to Evaluating Product Liability Cases %P 627-659 %A Edwina Rissland %T Argument Moves and Hypotheticals %P 129-143 %A Dean Schlobohm %T TA -- A Prolog Program which Analyzes Income Tax Issues under Section 318(A) of the Internal Revenue Code %P 765-815 %A Franciszek Studnicki %T Computational Aspects of Legal Interpretation %P 157-161 %A Charles Walter %T Introduction %P 1-4 %A Charles Walter %A Michael Parks %T Natural Models of Intelligence %P 5-27 %A Marshall Willick %T Professional Malpractice and the Unauthorized Practice of Professions: Some Legal and Ethical Aspects of the use of Computers and Decision-Aids %P 817-863 %A Gian Piero Zarri %T Inference Techniques for Intelligent Information Retrieval %P 247-274 - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - George R. Cross cross@wsu.CSNET Computer Science Department cross%wsu@csnet-relay.ARPA Washington State University faccross@wsuvm1.BITNET Pullman, WA 99164-1210 Phone: 509-335-6319 or 509-335-6636 ------------------------------ Date: Wed, 11 Dec 85 10:50:41 EST From: "Lisa F. Melcher" Subject: Seminar - Parallel Depth-First Search (MIT) DATE: Friday, December 13, 1985 TIME: 12:45.....Refreshments 1:00.....Lecture PLACE: NE43 - 512A "A PARALLEL ALGORITHM FOR DEPTH FIRST SEARCH" Richard Anderson MSRI A new parallel algorithm for constructing a depth first search tree in an undirected graph will be described. The algorithm is a P-RAM algorithm and uses several probabilistic algorithms as sub-routines. The run time of the algorithm is 2 sq.rt. log n. This makes it an almost RNC eps. algorithm, since the run time is less than n for any eps.>0. The standard sequential algorithm for depth first search can be shown to be "inherently sequential", so this shows that substantial speed up for depth first search is possible when a different approach is taken. David Shmoys Host ------------------------------ Date: Tue, 10 Dec 85 12:57:28 cst From: Ted Briggs Subject: Seminar - Quantitative Operators in Design (UTexas) Quantitative Operators in Design by David R Throop noon Friday Dec. 13 PAI 3.38 Doing design problems involves reasoning about quantities. AI methods have not made great headway into engineering design problems so far. They have lacked the power to reason about quantities. Meanwhile, other numeric techniques have lacked the AI systems' abilities in handling heuristic knowledge. I propose to extend AI representations to allow reasoning about quantities in design. My approach has these features: Encoding heuristic knowledge as frames with quantita- tive operators, operators with an action and a strength. Beginning design by solving subproblems, guided by the arbitration of conflicting quantitative heuristics. Reasoning quantitatively about the aggregate design, modifying it using non-local knowledge. Encoding the procedures to arbitrate conflicting heuristics as knowledge at a general level: the para- digm level. To test the power of this approach, I will apply it to an outstanding problem in Chemical Engineering: the synthesis of heat exchanger networks. This problem has many published non-AI computer approaches. This literature provides a test of my representation and its adequacy for real design prob- lems. ------------------------------ Date: 6 Dec 1985 1331-EST From: David A. Evans Subject: Seminar - An Object Model of Information (CMU) Speaker: David Beech, Center for Integrated Systems, Stanford University, and Hewlett-Packard Laboratories, Palo Alto Time: Wednesday, 11 December 1985, 1:30-3:00 PM, Wean 5409 Title: Towards an Object Model of the Representation and Use of Information Future general-purpose information systems will need to deal with a wide range of information, and offer flexible access to it, if they are to appeal to the potential millions of non-specialist users. For example, they should process pictures and sounds as naturaly as numbers and texts; they should answer questions which require some deduction from the often incomplete information previously given to the system; and they should move towards the support of natural language interfaces, including spoken inputs. An object-oriented model of the representation and use of information is proposed, with the necessary generality for the desription and design of such systems. Fundamental concepts including those of agent, object, type, action, formula, process, transaction, predicator and generator are intro- duced. Recursive functions, predicate calculus, and n-ary relations are brought together in a data abstraction framework, with an emphasis on intensional definition of concepts and their instantiation by means of predicators and generators. ------------------------------ Date: Fri, 6 Dec 1985 10:21 EST From: ELIZABETH%MIT-OZ@MIT-MC.ARPA Subject: Seminar - Parallel Networks that Learn Representations (MIT) MASSIVELY PARALLEL NETWORKS THAT LEARN REPRESENTATIONS Geoffrey Hinton Carnegie-Mellon University Monday, December 9 Eighth Floor Playroom, AI Lab 4:00 pm I shall describe a new learning procedure for massively parallel networks of neuron-like processing elements. The procedure adjusts the connection strengths in a multi-layered network so as to make it give the correct output vector when given an input vector. The units in the intermediate layers come to represent important implicit features of the task domain that generates the pairs of input/output vectors. As a result, the network can generalise appropriately to new cases. I shall describe a pattern recognition example in which the network constructs a balanced ecology of feature detectors, and a higher level task in which the network learns a set of relationships. The second example illustrates the ability of the network to recognize isomorphisms and make use of them in encoding and generalizing knowledge. ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Fri Dec 20 01:50:33 1985 Date: Fri, 20 Dec 85 01:50:20 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a025923; 15 Dec 85 14:51 EST Date: Sun 15 Dec 1985 11:38-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #186 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Wed, 18 Dec 85 15:38 EST AIList Digest Sunday, 15 Dec 1985 Volume 3 : Issue 186 Today's Topics: Seminars - Learning Objects from Images (MIT) & SDI Debate (SU) & Deduction as a Programming Methodology (UTexas) & Typed Equational Logic Programming (UTexas), Conference - Expert Systems in Government & Object-Oriented Programming System, Languages, and Applications ---------------------------------------------------------------------- Date: Tue, 10 Dec 85 20:06 EST From: Brian C. Williams Subject: Seminar - Learning Objects from Images (MIT) [Forwarded from the MIT bboard by SASW@MIT-MC.] Thursday , December 12 4:00pm Room: NE43- 8th floor Playroom The Artificial Intelligence Lab Revolving Seminar Series "Learning Symbolic Object Models From Images" Jonathan Connell MIT, AI Lab This talk will present the results of an implemented system for learning structural prototypes of objects directly from gray-scale images. The vision component of this system employs Brady's Smoothed Local Symmetries to divide an object into parts which are then described symbolically. The learning component takes these descriptions and forms a model of the examples presented in a manner similar to Winston's ANALOGY program. The problem of matching complex structured descriptions and the difficult task of reasoning about function from form will also be briefly discussed. ------------------------------ Date: Fri 13 Dec 85 17:56:01-PST From: Joan Feigenbaum Subject: Seminar - SDI Debate (SU) ``SDI: How Feasible, How Useful, How Robust?'' This will be a technical debate, covering both hardware and software aspects of SDI. Sponsor: Stanford Computer Science Department Date: December 19, 1985 Time: 8:00 p.m. Place: Terman Auditorium Organizer: Barbara Simons, IBM-SJ Moderator: Dr. Marvin L. Goldberger, President of Cal Tech. Former member of President's Science Advisory Committee and Consultant on Arms Control and International Security. Panelists: Advocates: Professor Richard Lipton, Professor of Computer Science at Princeton University, Current member of SDIO's Panel on Computing and Support of Battle Management. Major Simon Peter Warden, the Special Assistant to the Director of the SDIO and Technical Advisor to the Nuclear and Space Arms Talk with the USSR in Geneva. Opponents: Dr. Richard L. Garwin, IBM Fellow and Adjunct Professor of Physics at Columbia University, Physicist and Defense Consultant. Professor David Parnas, Lansdown Professor of Computer Science at the University of Victoria, Former member of the SDI Organization's Panel on Computing and Support of Battle Management. ------------------------------ Date: Fri 13 Dec 85 11:56:26-CST From: Subject: Seminar - Deduction as a Programming Methodology (UTexas) Gerard Huet INRIA (on Sabbatical at CMU) MCC-AI Lecture Thursday, December 19 at 10:00am Echelon I, Room 409 "Intuitionistic Higher-Order Natural Deduction as a Programming Methodology" The talk will review various aspects of the correspondence between types and propositions and its applications to type-checking and program proving. A higher-order intuitionistic natural deduction formalism, the Calculus of Constructions, is introduced and motivated by examples. ------------------------------ Date: Fri 13 Dec 85 11:55:40-CST From: Subject: Seminar - Typed Equational Logic Programming (UTexas) Typed Equational Logic as a Programming Language Gert Smolka Cornell University MCC-AI Lecture WEDNESDAY, DECEMBER 18 at 3:00pm Echelon I, Room 409 I will present a logic programming notation called TEL, which is based on typed equational logic. TEL avoids many of Prolog's difficulties since it has equality and functions, an expressive type system, and more satisfying control features. Furthermore, employing equational rather than relational logic supports term rewriting in addition to logic programming a la Prolog. TEL's major innovations come with its type system: * Both type containment (subtypes as in OBJ2) and parametric polymorphism (as in ML) are available. This requires type checking algorithms that solve inequalities. * Although, in general, types are computationally significant, every program can be automatically translated into an equivalent TEL program in which all type information is redundant. Thus TEL can be implemented with untyped narrowing. * Type declarations in TEL contain control information that enables a compiler to decide whether a function can be implemented by rewriting instead of full narrowing. This is a crucial optimization since rewriting requires neither backtracking nor unification. * The control information in TEL's type declarations also facilitates the exploitation of and-parallism. A straight- forward compile-time analysis can determine most subterms that will not share variables at run-time. ------------------------------ Date: 13 Dec 85 10:30:25 EST (Fri) From: Duke Briscoe Subject: Conference - Expert Systems in Government CALL FOR PAPERS THE SECOND ANNUAL CONFERENCE ON EXPERT SYSTEMS IN GOVERNMENT Tyson's Westpark Hotel, McLean, VA in suburban Washington, D.C. October 20 - 24, 1986 The conference is sponsored by the IEEE Computer Society and the Mitre Corporation in cooperation with AIAA/NCS. The objective of the conference is to explore the following: - knowledge based applications and supporting technologies - implementation and impact of emerging application areas - future trends in available systems and required research Classified and unclassified papers which relate to the use of knowledge based systems are solicited. The topics of interest include, but are not limited to, the following applications: Professional: engineering, finance, law, management, medicine Office Automation: text understanding, intelligent DBMS, intelligent systems Command & Control: intelligence analysis, planning, targeting, communications, air traffic control, battle management Exploration: outer space, prospecting, archaeology Weapon Systems: adaptive control, electronic warfare, Star Wars, target identification Equipment: CAD/CAM, design monitoring, maintenance, repair Software: automatic programming, maintenance, verification and validation Architecture: distributed knowledge based systems, parallel computing Project Management: planning, scheduling, control Education: concept formation, tutoring, testing, diagnosis Imagery: photo interpretation, mapping Systems Engineering: requirements, preliminary design, critical design, testing, quality assurance Tools and Techniques: PROLOG, knowledge acquisition and representation, uncertainty management Plant and Factory Automation Space Station Systems Human-Machine Interface Speech and Natural Language The program will consist of submitted and invited papers, which will provide an overview of selected areas. Contributed papers should be consistent with the following outline: 1. Introduction- state clearly the purpose of the work 2. Description of the actual work- must be new and significant 3. Results- discuss their significance 4. References Completed papers are to be no longer than 20 pages, including graphics. Four copies of the complete paper are to be submitted to: Dr. Kamal Karna, Conference Chairman IEEE Computer Society 1730 Massachusetts Ave., NW Washington, D.C. 20036-1903 Author's Schedule: Four copies of manuscript May 1, 1986 Acceptance letter June 15, 1986 Camera-ready copy July 15, 1986 Conference Chairman: Dr. Kamal Karna Washington AI Center Mitre Corporation karna@mitre Program Committee: Co-chairman: Classified and Unclassified: Dr. Richard Martin Associate Director, Software Engineering Institute Carnegie Mellon University AIMARTN@MCC Co-chairman: Unclassified Dr. Kamran Parsaye President Intelliware, Inc. ------------------------------ Date: Fri, 13 Dec 85 15:48:07 PST From: Bay.pa@Xerox.ARPA Reply-to: Ingalls%Apple.csnet@CSNET-RELAY.ARPA Subject: Conference - Call for Papers, OOPSLA86 Call For Papers and Participation ACM Conference on Object Oriented Programming Systems, Languages, and Applications September 29 - October 2, 1986, Portland, Oregon OOPSLA-86 is a new ACM-sponsored conference that brings together users and implementors of object oriented systems. Through tutorials, papers, panel discussions and workshops, as well as demonstrations, exhibits and videotapes, OOPSLA-86 will provide a forum for sharing experience and knowledge among experts and novices alike. We invite technical papers, case studies, and surveys in the following areas: Theory: Including core definition of object oriented programming, semantic models and methodology. Languages: Existing object oriented languages, extensions to conventional languages, and new languages. Implementation: Including architectural support, compilation and interpretation, and special techniques. Tools and Environments: Including user interfaces, utilities and operating system support. Applications: Commercial, educational, and scientific applications that exploit object oriented programming. Related Work: The object oriented paradigm in other fields such as databases and operating systems. Papers on other relevant topics are welcome, as are proposals for workshops and panel discussions. All papers will be refereed prior to selection and inclusion in the conference proceedings. Technical papers will be selected on the basis of originality and contribution to the state of the art of design, implementation, methodology, or practice. Survey papers will be selected on the basis of how well they crystallize and integrate, in a way not previously presented, knowledge about one or more aspects of the field. Papers must be submitted in English, and should be no longer than 25 double-spaced pages. The cover page should include a title, an abstract of not more than 100 words, and author's name, affiliation, address and phone number. Five copies must be received by the Program Chairman at the address below, no later than April 1, 1986. Authors will be notified of acceptance by May 1, 1986, and final versions of accepted papers will be due by June 15, 1986. As the proceedings of this conference will be widely disseminated, publication of more than an abstract of a submitted paper is likely to inhibit republication in ACM's refereed publications. A room at the conference will be reserved for video presentations that illustrate or supplement the concepts conveyed in other presentations. Submissions must run no longer than 15 minutes, and should be on 3/4-inch U-Matic format tape. Tapes must be received by the Video Chairman at the address below, no later than July 1, 1986. Conference Chairmen Daniel Bobrow (Xerox PARC) (bobrow.pa@XEROX.ARPA) Alan Purdy (Servio Logic Development) Program Chairman Daniel Ingalls, MS 22-Y (ingalls%apple.csnet@CSNET-RELAY) Apple Computer 20525 Mariani Ave. Cupertino, CA 95014 Video Chairman David Robson (robson.pa@XEROX.ARPA) Xerox PARC 3333 Coyote Hill Road Palo Alto, CA 94304 ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Wed Jan 1 12:37:37 1986 Date: Wed, 1 Jan 86 12:37:27 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a026119; 15 Dec 85 15:04 EST Date: Sun 15 Dec 1985 11:47-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #187 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Wed, 18 Dec 85 15:41 EST AIList Digest Sunday, 15 Dec 1985 Volume 3 : Issue 187 Today's Topics: Humor - Alice's LISP Machine ---------------------------------------------------------------------- Date: Thu 5 Dec 85 11:25:15-CST From: AI.HASSAN@MCC.ARPA Subject: Humor - Thanksgiving Story [Forwarded from NIKHIL@MIT-XX] [Forwarded from the UTexas bboard by Laws@SRI-AI.] ALICE'S LISP MACHINE Chris Stacy, Alan Wecsler, and Noel Chiappa This song is called "MIT's AI Lab". It's about MIT and the AI Lab, but "MIT's AI Lab" is not the name of the lab, that's just the name of the song. That's why I call the song "MIT's AI Lab." Now it all started two full dumps ago, on Thanksgiving, when my friend and I went up to visit the hackers at AI lab on the ninth floor. But the hackers don't always live on the ninth floor, they just go there to use these complex order code stack machines they call Lisp Machines. And using a special purpose processor like that, they got a lot of room upstairs where DDT used to be, and havin' all that ROOM they decided that they didn't have to collect any garbage for a long time. We JFCLed up here and found all the garbage in there and we decided that it'd be a friendly gesture for us to take all the garbage down to the system dump. So we took the half-a-meg of garbage, put it in the back of a red ECL Multibus, took subrs and hacks and implementations of defstruction, and headed on toward the system dump. Well, we got there and there was a big pop up window and a write protect across the dump sayin', "This Garbage Collecter Under Development on Thanksgiving," and we'd never heard of a garbage collector NOP'd out on Thanksgiving before, and with tears in our eyes, we CDR'd off into the sunset lookin' for another place to put the garbage. We didn't find one 'til we came to a side area, and off the side of the side area was three hundred megabyte disk, and in the middle of the disk was another heap of garbage. And we decided that one big heap was better than two little heaps, and rather than page that one in, we decided to write ours out. That's what we did. Branched back to the Lisp Listener, had a Chinese Thanksgiving dinner that couldn't be beat, went to SI:PROCESS-WAIT SLEEP, and didn't get up until the next quantum, when we got a funcall from Mr. Greenblatt. He said, "Kid, we found your name on a cons at the bottom of a half-a-meg of garbage and I just wanted to know if you had any information about it". And I said, "Yes sir, Mr. Greenblatt, I cannot tell a lie. I put that structure under that garbage." After speakin' to Greenblatt for about forty-five million clock ticks on the telnet stream, we finally arrived at the truth of the matter and he said that we had to go down and link up the garbage, and also had to go down and speak to him at the Lisp Machine Factory. So we got in the red ECL Multibus with the subrs and hacks and implementations of defstruction and headed on toward the Lisp Machine Factory. Now, friends, there was only one of two things that Greenblatt could've done at the Lisp Machine Factory, and the first was that he could've given us another 64K board for bein' so brave and honest on BUG-LISPM (which wasn't very likely, and we didn't expect it), and the other thing was that he could've flamed at us and told us never to be seen BLTing garbage around in the vicinity again, which is what we expected. But when we got to the Lisp Machine Factory, there was a third COND-clause that we hadn't even counted upon, and we was both immediately Process-Arrested, Deexposed, and I said, "Greenblatt, I can't GC up the garbage with these here ARREST-REASONS on". He said: "Output-Hold, kid, and get in the back of the Control CAR." ...And that's what we did...sat in the back of the Control CAR, and drove to the sharpsign quote open scene-of-the-crime close. I wanna tell you 'bout the town of Cambridge, Massachusetts, where this is happenin'. They got seven hunnert stop signs, no turn on red, and two campus police CARs, but when we got to the sharpsign-quote-open scene-of-the-crime close, there was five Lisp Machine hackers and three scope carts, bein' the biggest hack of the last ten years and everybody wanted to get in the HUMAN-NETS story about it. And they was usin' up all kinds of digital equipment that they had hangin' around the Lisp Machine Factory. They was takin' backtraces, stack traces, plastic wire wraps, blueprints, and microcode loads...And they made seventeen 1K-by-32 pixel multi-flavored windows with turds and arrows and a scroll bar on the side of each one with documentation panes explainin' what each one was, to be used as evidence against us. ...Took pictures of the labels, blinkers, the cursors, the pop up notification windows, the upper right corner, the lower left corner ...and that's not to mention the XGP'd screen images! After the ordeal, we went back to the Factory. Greenblatt said he was gonna locate us in a cell. He said: "Kid, I'm gonna INTERN you in a cell. I want your manual and your mouse." I said, "Greenblatt, I can understand your wantin' my manual, so I don't have any documentation about the cell, but what do you want my mouse for?" and he said, "Kid, we don't want any window system problems". I said, "Greenblatt, did you think I was gonna deexpose myself for litterin'?" Greenblatt said he was makin' sure, and, friends, Greenblatt was, 'cause he took out the left Meta-key so I couldn't double bucky the rubout and cold-boot, and he took out the Inspector so I couldn't click-left on Modify, set the PROCESS-WARM-BOOT-ACTION on the window, *THROW around the UNWIND-PROTECT and have an escape. Greenblatt was makin' sure. It was about four or five hours later that Moon--(remember Moon? This here's not a song about Moon)-- Moon came by and, with a few nasty sends to Greenblatt on the side, bailed us out of core, and we went up to the Loft, had another Chinese dinner that couldn't be beat, and didn't get up until the next evening, when we all had to go to court. We walked in, sat down, Greenblatt came in with the seventeen 1K-by-32 pixel multi-flavored windows with turds and arrows and documentation panes, sat down. McMahon came in, said, "All rise!" We all stood up, and Greenblatt stood up with the seventeen 1K-by-32 pixel multi-flavored windows with turds and arrows and documentation panes, and the judge walked in, with an LA36, and he sat down. We sat down. Greenblatt looked at the LA36... then at the seventeen multi flavored windows with the turds and arrows and documentation panes... and looked at the LA36... and then at the seventeen 1K-by-32 pixel multi-flavored windows with turds and arrows and documentation panes, and began to cry. Because Greenblatt came to the realization that it was a typical case of LCS state-of-the-art technology, and there wasn't nothin' he could do about it, and the judge wasn't gonna look at the seventeen 1K-by-32 pixel multi-flavored windows with turds and arrows and documentation panes, explainin' what each one was, to be used as evidence against us. And we was fined fifty zorkmids and had to rebuild the world load...in the snow. But that's not what I'm here to tell you about. I'm here to talk about the Lab. ______________________________________________________________________ They got a buildin' down in Cambridge called Technology Square, where you walk in, you get your windows Inspected, detected, neglected and Selected! I went down and got my interview one day, and I walked in, sat down (slept on the beanbag in 926 the night before, so I looked and felt my best when I went in that morning, 'cause I wanted to look like the All-American High School Tourist from Sunnyvale. I wanted to feel like ... I wanted to be the All-American Kid from Sunnyvale), and I walked in, sat down, I was gunned down, brung down, locked out and all kinds of mean, nasty, ugly things. And I walked in, I sat down, KAREN gave me a piece of paper that said: "Kid, see the CLU hackers on XX." I went up there, I said, "Eliot, I wanna lose. I wanna lose! I wanna see hacks and kludges and unbound variables and cruft in my code! Eat dead power supplies with cables between my teeth! I mean lose! lose! lose!" And I started jumpin' up and down, yellin' "LOSE! LOSE! LOSE!" and Stallman walked in and started jumpin' up and down with me, and we was both jumpin' up and down, yellin', "LOSE! LOSE! LOSE! LOSE!!" and some professor came over, gave me a 6-3 degree, sent me down the hall, said "You're our distinguished lecturer." Didn't feel too good about it. Proceeded down the infinite corridor, gettin' more inspections, rejections (this IS MIT), detections, neglections, and all kinds of stuff that they was doin' to me there, and I was there for two years... three years... four years... I was there for a long time goin' through all kinds of mean, nasty, kludgy things, and I was havin' a tough time there, and they was inspectin', injectin', every single part of me, and they was leavin' no part unbound! ______________________________________________________________________ Proceeded through, and I finally came to see the very last man. I walked in, sat down, after a whole big thing there. I walked up, and he said, "Kid, we only got one question: Have you ever been arrested"? And I proceeded to tell him the story of the half-a-meg of garbage with full orchestration and five-part harmony and stuff like that, and other phenomenon. He stopped me right there and said, "Kid, have you ever been to court"? And I proceeded to tell him the story of the seventeen 1K-by-32 pixel multi-flavored windows with turds and arrows and documentation panes... He stopped me right there and said, "Kid, I want you to go over and sit down on that bench that says 'LISP Machine Group'... NOW, KID!" And I walked over to the bench there, and there's... The LISP Machine Group is where they put you if you may not be moral enough to join Symbolics after creatin' your special form. There was all kinds of mean, nasty, ugly-lookin' people on the bench there ... there was Microcoders, DPL hackers, File System hackers, and Window System Hackers!! Window System hacker sittin' right there on the bench next to me! And the meanest, ugliest, nastiest one... the kludgiest Window System hacker of them all... was comin' over to me, and he was mean and ugly and nasty and horrible and all kinds of things, and he sat down next to me. He said, "Kid, you get a new copy of the sources?" I said, "I didn't get nothin'. I had to rebuild the world load." He said, "What were you arrested for, kid?" and I said, "Littering..." And they all moved away from me on the bench there, with the hairy eyeball and all kinds of mean, nasty things, 'til I said, "And making gratuitous modifications to LMIO; sources..." And they all came back, shook my hand, and we had a great time on the bench talkin' about microcoding, DPL designing, file-system hacking, ... and all kinds of groovy things that we was talkin' about on the bench, and everything was fine. We was drinking Coke smoking all kinds of things, until the RA came over, had some paper in his hand, held it up and said: "KIDS-THIS-EXAM-S-GOT-FOURTY SEVEN-WORDS-THIRTY-SEVEN-MULTIPLE-CHOICE-QUESTIONS FIFTY-EIGHT-WORDS-WE-WANT-TO-KNOW-THE-DETAILS OF-THE-HACK-THE-TIME-OF-THE-HACK-AND-ANY OTHER-KIND-OF-THING-YOU-GOT-TO-SAY PERTAINING-TO-AND-ABOUT-THE-HACK-ANY-OTHER KIND-OF-THING-YOU-GOT-TO-SAY-WE-WANT-TO-KNOW THE-ARRESTED-PROCESS'-NAME-AND-ANY OTHER-KIND-OF-THING..." And he talked for forty-five minutes and nobody understood a word that he said. But we had fun rolling the mice around and clickin' on the buttons. I filled out the special form with the four-level macro defining macros. Typed it in there just like it was and everything was fine. And I put down my keyboard, and I switched buffers, and there ... in the other buffer... centered in the other buffer... away from everything else in the buffer... in parentheses, capital letters, backquotated, in 43VXMS, read the following words: "Kid, have you featurized yourself"? I went over to the RA. Said, "Mister, you got a lot of damned gall to ask me if I've featurized myself! I mean, I mean, I mean that you send, I'm sittin' here on the bench, I mean I'm sittin' here on the Lisp Machine Group bench, 'cause you want to know if I'm losing enough to join the Lab, burn PROMs, power supplies, and documentation, after bein' on SF-LOVERS?" ______________________________________________________________________ He looked at me and said, "Kid, we don't like your kind! We're gonna send your user-id off to the DCA in Washington"! And, friends, somewhere in Washington, enshrined on some little floppy disk, is a study in ones and zeros of my brain-damaged programming style... ______________________________________________________________________ And the only reason I'm singin' you the song now is 'cause you may know somebody in a similar situation. Or you may be in a similar situation, and if you're in a situation like that, there's only one thing you can do: [ CHORUS ] You know, if one person, just one person, does it, they may think he's really dangerous and they won't take him. And if two people do it, in harmony, they may think they're both LISP hackers and they won't take either of them. And if three people do it! Can you imagine three people walkin' in, singin' a bar of "MIT's AI Lab" and walkin' out? They may think it's an re-implementation of the window system! And can you imagine fifty people a day? I said FIFTY people a day, walkin' in, singin' a bar of "MIT's AI Lab" and walkin' out? Friends, they may think it's a MOVEMENT, and that's what it is: THE MIT AI LAB ANTI-LOSSAGE MOVEMENT! And all you gotta do to join is to sing it the next time it comes around on the circular buffer. With feelin'. You can hack anything you want on MIT Lisp Machines You can hack anything you want on MIT Lisp Machines Walk right in and begin to hack Just push your stuff right onto the stack You can hack anything you want on MIT Lisp Machines (but don't forget to fix the bug...on MIT Lisp Machines!) ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Fri Dec 20 01:50:14 1985 Date: Fri, 20 Dec 85 01:49:59 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a005116; 16 Dec 85 12:23 EST Date: Mon 16 Dec 1985 09:11-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #188 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Wed, 18 Dec 85 15:57 EST AIList Digest Monday, 16 Dec 1985 Volume 3 : Issue 188 Today's Topics: Query - Superset of Common LISP on a Chip, Logic Programming - LIPS, Policy - ADS Discussion, Literature - Correction: Law and Technology Bibliography, Logic - Factual Counterfactuals ---------------------------------------------------------------------- Date: Fri 13 Dec 85 11:52:12-PST From: Ali Ozer Subject: Superset of Common LISP... [Forwarded from the Stanford bboard by Laws@SRI-AI.] ...on a chip? I just read in this month's Byte (Dec 1985, p10): "Texas Instruments is developing a 32-bit CMOS LISP processor chip... Roughly 10 times more complex than a 68000, the 40-MHz processor will directly execute a superset of Common LISP with extensions like object-oriented programming and message-passing..." This sounds truly amazing to me. A superset of what is described in Steele's Common Lisp manual? That would be a *powerful* chip. Has anyone read more about this? (All Byte had was the above sentence...) Ali ------------------------------ Date: Fri, 13 Dec 85 10:36 EST From: Tim Finin Subject: LIPS For a good discussion of the systematic benchmarking and analysis of a Prolog implementation, see "THE PRODUCTION AND EVALUATION OF A SET OF PROLOG BENCHMARKS" by Paul F. Wilk, Dept. of Artificial Intelligence, University of Edinburgh. I've cliped the following from this paper: The program "Nrev" is seen as a standard measurement for benchmarking Pro- log implementations. This is a simple symbol-crunching program for the naive reversal of a list (see Appendix B). It seems fairer to use this program as a standard measure rather than any program involving arithmetic; which favours the DEC-10 compiler. The following table shows the estimated Logical Inferences Per Second (LIPS) for these Prolog system configura- tions. Nrev performs 496 logical inferences (procedure calls) each time it is executed. Therefore, dividing the number of logical inferences for one execution by the execution time, in seconds, gives the number of LIPS. |===============|=======|=======|=======|=======|=======|=======|=======| |Implementation |PrologC|PrologI|Cprolog|Cprolog|PDP-11 |POPLOG |York | |---------------|-------|-------|-------|-------|-------|-------|-------| |System range |DEC-10 |DEC-10 |VAX-11 |ICLPerq|PDP-11 |VAX-11 |ICLPerq| |---------------|-------|-------|-------|-------|-------|-------|-------| |Processor |KL-10 |KL-10 |11-750 |2910 |11-60 |11-780 |2910 | |---------------|-------|-------|-------|-------|-------|-------|-------| |Translator |Compil.|Interp.|Interp.|Interp.|Interp.|Compil.|Interp.| |---------------|-------|-------|-------|-------|-------|-------|-------| |LIPS (Nrev) | 45592| 2385| 800| 484| 888| 1858| 56| |---------------|-------|-------|-------|-------|-------|-------|-------| |Cputime ratio | 1 | 19.12| 56.99| 94.20| 51.34| 24.54| 814.14| |===============|=======|=======|=======|=======|=======|=======|=======| Table 3. Nrev: LIPS comparison. The paper discusses measuring other attributes (e.g. memory utilization) and provides a variety of benchmarks to measure particular language features (e.g. time to assert a unit clause). I've checked several implementations/machines here. For CProlog, I recall measures like 2400 LIPS for a 785, 400 LIPS for a HP9836, and 140 LIPS for a MacIntosh. Our Symbolics-prolog (compiled) tested out at about 40K LIPS.Here is the program I've used to do the benchmarking. I think it comes from Paul Wilk. % lips runs the nrev benchmark 10 times and writes the estimated lips. lips :- lips(10,L), nl,write(L),write(' lips'). % Lips(N,L) runs the nrev benchmark N times and binds L to % an estimate of the LIPS. lips(N,L) :- init(X), T1 is cputime, tests(N,X), T2 is cputime, control(N,X), T3 is cputime, Testtime is T2-T1, Overhead is T3-T2, Nettime is Testtime-Overhead, write(' '), write(Testtime-Overhead=Nettime), nl, L is (496*N)/Nettime. tests(N,X) :- from(1,N,I), test(X), fail. tests(N,X). control(N,X) :- from(1,N,I), dummy(X), fail. control(N,X). from(I,I,I) :- !. from(L,N,I) :- N1 is (L+N)//2, from(L,N1,I). from(L,N,I) :- L1 is (L+N)//2+1, from(L1,N,I). %NREV: Naive Reverse of 30 element list init([1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20, 21,22,23,24,25,26,27,28,29,30]). dummy(L). test(L0) :- nreverse(L0,L1). nreverse([X|L0],L) :- nreverse(L0,L1), concatenate(L1,[X],L). nreverse([],[]). concatenate([X|L1],L2,[X|L3]) :- concatenate(L1,L2,L3). concatenate([],L,L). ------------------------------ Date: Fri 13 Dec 85 17:29:27-PST From: LOUROBINSON@SRI-AI.ARPA Subject: For AIList Ken: (For AIList:) In response to AI.Duffy at UTexas and his message regarding the AI&DS name change: Why the cheap shot? It's obvious why the name change was necessary. Why exploit them for the sake of Duffy's unsigned, heavy-handed cleverness? Lou Robinson ------------------------------ Date: Sun, 15 Dec 1985 13:23 PLT From: George Cross Subject: Correction: Law and Technology Bibliography In my posting of the table of contents of the Proceedings of the First Annual Law and Technology Conference (Ailist V3 #117, 12/13/85), I left out Anne Gardner's paper and mispaged Zarri's: %T Overview of an Artificial Intelligence Approach to Legal Reasoning %A Anne v.d.L. Gardner %B Computing Power and Legal Reasoning %E Charles Walter %I West Publishing Company %P 247-274 %D 1985 %C St. Paul %X ISBN 0-314-95570-4 %T Inference Techniques for Intelligent Information Retrieval %A Gian Piero Zarri %B Computing Power and Legal Reasoning %E Charles Walter %I West Publishing Company %P 215-246 %D 1985 %C St. Paul %X ISBN 0-314-95570-4 ---- George - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - George R. Cross cross@wsu.CSNET Computer Science Department cross%wsu@csnet-relay.ARPA Washington State University faccross@wsuvm1.BITNET Pullman, WA 99164-1210 (509)-335-6319/6636 Acknowledge-To: George Cross ------------------------------ Date: 13 Dec 85 1143 PST From: Matthew Ginsberg Subject: interesting counterfactual > From: Mike Dante > . . . > (0) Suppose a class consists of three people, a 6 ft boy (Tom), a 5 ft girl > (Jane), and a 4 ft boy (John). Do you believe the following statements? > > (1) If the tallest person in the class is a boy, then if the tallest > is not Tom, then the tallest will be John. > (2) A boy is the tallest person in the class. > (3) If the tallest person in the class is not Tom then the tallest > person in the class will be John. > > How many readers believe (1) and (2) imply the truth of (3)? It seems to me that this example gives insight as to what the status is of a counterfactual whose premise is true. The general view among philosophers on this is that the truth or falsity of a counterfactual with a true premise depends simply on the truth or falsity of its conclusion, but this example seems to run against this view. The reason seems to me to be that the conclusion should be investigated as if the premise had read, "If the tallest person in the class is *necessarily* a boy." In other words, in constructing possible worlds in which "the tallest is not Tom" (i.e., in investigating the truth of the conclusion of (1)), possible worlds in which the tallest is not a boy are disallowed. Thus (1) and (2) can be true while (3) is false. I don't know what to make of this generally. It appears that the problem can only arise with enbedded counterfactuals; perhaps it is reasonable to treat them specially, rewriting (1) as: If the tallest person in the class is a boy and the tallest is not Tom, then the tallest will be John. The two approaches are equivalent, but I fear there may be substantial ramifications to messing around with the semantics in this fashion. Matt Ginsberg ------------------------------ End of AIList Digest ******************** From comsat@vtcs1 Fri Dec 20 01:49:52 1985 Date: Fri, 20 Dec 85 01:49:40 est From: comsat@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a028378; 18 Dec 85 12:51 EST Date: Wed 18 Dec 1985 09:17-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #189 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Wed, 18 Dec 85 16:20 EST AIList Digest Wednesday, 18 Dec 1985 Volume 3 : Issue 189 Today's Topics: Query - Finding Loops in Prolog & CAI, AI Tools - Object oriented programming in Common Lisp, Policy - ADS Message & Advertisements, Logic - Counterfactuals ---------------------------------------------------------------------- Date: Tue, 17 Dec 85 15:34:54 est From: Catherine A. Meadows Subject: loops Does anyone out there know of any work that has been done on checking for loops in Prolog (other than the papers recently published in ACM Sigplan)? Cathy Meadows meadows@nrl-css ------------------------------ Date: 17 Dec 85 23:38 EST From: Gunther @ DCA-EMS Subject: REQUEST FOR INFO ON CAI ORIENTED SHELLS IF ANYONE IS WORKING IN THE AREA OF EXPERT CAI DEVELOPMENT SYSTEMS, I'D LIKE SOME INFORMATION ON APPROACHES, OTHER WORK, PACKAGES, ETC ... THANKS, J.L. FEINSTEIN [Contact AI-Ed-Request@SUMEX-AIM.ARPA to get in touch with a like-minded group. -- KIL] ------------------------------ Date: 17 DEC 85 13:32-N From: DESMEDT%HNYKUN52.BITNET@WISCVM.WISC.EDU Subject: Object oriented programming in Common Lisp In reply to the following request: From: Nick Davies (at GEC Research) Subject: Object oriented programming in Common Lisp > Does anyone have or know of an implementation of Flavors or any other > object-oriented programming system in Common Lisp ? I would like to mention CORBIT, which is ORBIT rewritten in NIL Common Lisp at the University of Nijmegen (the Netherlands). CORBIT is an object oriented extension of Common Lisp which is NOT based on message-passing but on the idea of generic functions. An ultra-short example illustrating the difference: Flavors: (send window-1 :expose) CORBIT: (expose window-1) What is seen as a message in Flavors is a function in CORBIT. So it is not the object, but the message which is functional. This has a number of advantages. One advantage is that it is straightforward to use ordinary lambda-binding to localize inherited definitions. Another one is that the generic functions can be traced like any other Lisp function. The overall consequence is that CORBIT fits more neatly in the Lisp way of thinking than message-passing systems, is simpler to understand and implement, and yet offers the same possibilities. Koenraad De Smedt, DESMEDT@HNYKUN52 (bitnet) Psychological Laboratory University of Nijmegen The Netherlands [The recent Xerox PARC work on CommonLoops has a similar functional flavor -- as does ADA, of course. -- KIL] ------------------------------ Date: Mon 16 Dec 85 14:00:25-PST From: Wilkins Subject: AIDS Despite Duffy's bafflement, it seems obvious why they changed their name and it is important to the community to learn of such name changes. But, isn't there already someone named Advanced Decision Systems? Sounds familiar, but I cannot recall . . . ------------------------------ Date: Tue, 17 Dec 85 17:40:04 EST From: David_West%UMich-MTS.Mailnet@MIT-MULTICS.ARPA Subject: Policy You said in AIList v3 #185: "I [...] do screen out job ads [...]" For those of us out here in the boonies, the nets are the only available *current* source of information on what's really happening: the print media are agonizingly slow, particularly if one must rely on a university library. >From time to time, we boonie-dwellers even need to find jobs, and most AI jobs are outside the boonies. You didn't say whether you screen out job ads because of ARPAnet policy, or because they have a better home elsewhere. If it's the former, I think the policy is misguided, at least for academic/research jobs. If it's the latter, and there is in fact a network source for this kind of information, I don't know of it, and would like to. (I have seen *very* occasional academic fellowship announcements here: perhaps as many as two in the last year. Presumably there were more that went unreported.) [There are many reasons for the policy. (1) There are the Arpanet restrictions on commercial use: academic ads are generally considered acceptable but industrial ads are not. (I don't understand the distinction -- placement of personnel in the defense industry is in the interest of the net's military sponsors -- but rules are rules.) Even if Arpanet sponsors and host administrators did not object, a fair proportion of any such messages draw flak from other readers; I don't need the hassle of defending my editorial policies that often. (2) There is already a channel, Arpanet-BBoards@MIT-MC, for academic announcements and certain commercial announcements; I prefer to avoid duplicate traffic. (3) AIList carries enough traffic without adding another class of messages, particularly one that is of little interest to most readers and of no value in the historical archive. One difficulty in pruning the list is in screening out duplicates of messages submitted in previous months or years -- I have to pull the archives and scan for half-remembered occurrences. (4) I am unable to draw a clean distinction between messages of AI interest and those that are not. A search for a lab director or lecturer is of interest, but how about an ad for students to run an AI lab's computers? (4) A few of the people submitting ads lack perspective and are difficult to deal with -- any attempt to reject just some of the messages or to modify them to make them less commercial may lead to multiple interchanges with the authors. It is easier to reject such messages entirely. If someone else wants to moderate a list of placement ads or of resumes, he (or she) will probably find it acceptable to the net community. The volume of such solicitations put out by certain AI departments indicates a desire for such a service. (IEEE tried to serve this need among electrical engineers and computer scientists with their Professional Abstracts Registry; it couldn't be sustained, but the task has been passed on to a commercial information server.) I just find it better not to mix this traffic with AIList. I suggest that you check CACM and IEEE Computer for recent classified ads, or just contact companies listed in any recent AI magazine or conference proceedings. There are also some recruiters specializing in AI (Halbrect Associates, Inc.; JDG Associates, Ltd; Klein/Thaler Executive Search; Artificial Intelligence Referral Service; S.J. Parker and Associates); look for their ads in IEEE Spectrum and the AI magazines. -- KIL] ------------------------------ Date: 17 Dec 85 08:48:00 EDT From: "CUGINI, JOHN" Reply-to: "CUGINI, JOHN" Subject: counterfactuals >> From: Mike Dante >> >> (0) Suppose a class consists of three people, a 6 ft boy (Tom), >> a 5 ft girl (Jane), and a 4 ft boy (John). Do you believe the >> following statements? >> >> (1) If the tallest person in the class is a boy, then if the tallest >> is not Tom, then the tallest will be John. >> (2) A boy is the tallest person in the class. >> (3) If the tallest person in the class is not Tom then the tallest >> person in the class will be John. >> >> How many readers believe (1) and (2) imply the truth of (3)? > It seems to me that this example gives insight as to what the status > is of a counterfactual whose premise is true. The general view among > philosophers on this is that the truth or falsity of a counterfactual > with a true premise depends simply on the truth or falsity of its > conclusion, but this example seems to run against this view. > The reason seems to me to be that the conclusion should be investigated > as if the premise had read, "If the tallest person in the class is > *necessarily* a boy." In other words, in constructing possible worlds > in which "the tallest is not Tom" (i.e., in investigating the truth of the > conclusion of (1)), possible worlds in which the tallest is not a boy are > disallowed. Thus (1) and (2) can be true while (3) is false. I don't think there's a problem here. If we read the phrase "if the tallest is not Tom, then the tallest will be John" truth-functionally, in BOTH (1) and (3) then it's clearly true, since the antecedent is false - Tom is the tallest - and hence (1) and (3) are both true. If we read it counterfactually, eg: "what would happen if Tom shrank to a height of 3 angstroms, leaving only John and Jane? Would John be the tallest, true or false?" then it's clearly false in both cases, and so (1) and (3) are BOTH false. If we read (1) as: (1) If the tallest person in the class is a boy, then what would happen if Tom shrank to a height of 3 angstroms, leaving only John and Jane? Would John be the tallest, true or false? Modus ponens remains with its virtue intact, in either case. It's only when we read the phrase truth-functionally in (1), and counterfactually in (3) that an apparent conflict arises. Further confusing the issue is that (1) contains two implications, and the first sounds very truth-functional, leading you to read the second implication in the same way; but in isolation, in (3), it sounds counterfactual. If someone regards this as a real problem let him/her express it formally, and use different signs for different implication, eg "=>" for truth-functional, "->" for counterfactual. John Cugini ------------------------------ Date: 17 Dec 85 16:51:21 EST (Tue) From: Dana S. Nau Subject: Re: interesting counterfactual From: Matthew Ginsberg > From: Mike Dante > (0) Suppose a class consists of three people ... ... the conclusion should be investigated as if the premise had read, "If the tallest person in the class is *necessarily* a boy." In other words, in constructing possible worlds in which "the tallest is not Tom" (i.e., in investigating the truth of the conclusion of (1)), possible worlds in which the tallest is not a boy are disallowed. ... I don't think this solves the problem. Regardless of whether you allow such worlds or disallow them, you will still be allowing some worlds in which John is the tallest, and allowing others in which he is not. I don't know what to make of this generally. It appears that the problem can only arise with enbedded counterfactuals ... I disagree. In Dante's example, the APPARENT problem was that the strict logical interpretation of the statements didn't pin us down to the particular counterfactual world that we "obviously" wanted. But I think the real problem is more general. It's analogous to the frame problem: if we're going to deny some fact in order to create a counterfactual world, then what OTHER things are to be changed and what things are to remain the same? In general, it may not be clear which counterfactual world we want. In Dante's example above, the "obvious" counterfactual world was one in which Tom did not exist and all the other axioms were unchanged. But one could easily specify examples in which the removal of Tom would cause inconsistency unless some of the other axioms were changed too. In cases such as this, there may be many possible ways to change the axioms. It shouldn't be too difficult to construct an example in which different kinds of changes would seem best to different people--and thus different people would reach conflicting conclusions about what the changed world would be like. ------------------------------ End of AIList Digest ******************** From uucp Mon Feb 10 02:47 EST 1986 >From SRI-AI.ARPA!LAWS Mon Feb 10 02:24:41 1986 remote from seismo Received: by seismo.CSS.GOV; Mon, 10 Feb 86 02:24:41 EST Resent-Message-Id: <8602100724.AA00759@seismo.CSS.GOV> Message-Id: <8602100724.AA00759@seismo.CSS.GOV> Mail-From: LAWS created at 20-Dec-85 09:28:27 Date: Fri 20 Dec 1985 09:25-PST From: AIList Moderator Kenneth Laws Reply-To: seismo!SRI-AI.ARPA!AIList Us-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #190 To: SRI-AI.ARPA!AIList Resent-Date: Sun 9 Feb 86 22:26:35-PST Resent-From: Ken Laws Resent-To: vtisr1!fox Status: R AIList Digest Friday, 20 Dec 1985 Volume 3 : Issue 190 Today's Topics: Seminars - Example-Based Reasoning (UPenn) & Clausal Intuitionistic Logic (UPenn) & Robot Control with Kinesthetics (UPenn) & HORNLOG: Horn Clause Logic with Graph Rewriting (UPenn), Course - Deduction and Computation (CMU) & Symbolic Computation (UTexas) ---------------------------------------------------------------------- Date: Tue, 17 Dec 85 09:50 EST From: Tim Finin Subject: Seminar - Example-Based Reasoning (UPenn) 3pm Tuesday, December 17, 1985 Room 216 Moore School University Of Pennsylvania EXAMPLE-BASED REASONING EDWINA L. RISSLAND, HARVARD In this talk, I shall discuss example-based reasoning, particularly in the contexts of assisting in the preparation of legal arguments and offering on-line explanations. In the case of legal argumentation, I discuss how hypotheticals serve a central role in analyzing the issues in a case and describe a program, called HYPO, which generates legal hypotheticals, and an environment, called COUNSELOR, which provides support for legal reasoning and other strategic tasks, like resource management. I'll briefly describe our current work on on-line assistance and how we are trying to make it more intelligent by embedding custom-tailored examples in the explanations. I'll also discuss some general issues about examples such as their generation, structure and importance in reasoning, especially in the domains of mathematics and the law. ------------------------------ Date: Tue, 17 Dec 85 09:50 EST From: Tim Finin Subject: Seminar - Clausal Intuitionistic Logic (UPenn) 3pm Thursday, December 19, 1985 Room 216 Moore School University of Pennsylvania FIXED POINT SEMANTICS AND TABLEAU PROOF PROCEDURES FOR A CLAUSAL INTUITIONISTIC LOGIC L. THORNE MCCARTY COMPUTER SCIENCE DEPARTMENT RUTGERS UNIVERSITY Since the advent of Horn clause logic programming in the mid-1970's, there have been numerous attempts to extend the expressive power of Horn clause logic while preserving some of its attractive computational properties. This talk will present a clausal language which extends Horn clause logic by adding negations and embedded implications to the right hand side of a rule, and which interprets these new rules intuitionistically, in a set of partial models. The resulting system will be shown to have a fixed point semantics which resembles the fixed point semantics of Horn clauses, and a tableau proof procedure which generalizes Horn clause refutation proofs. Soundness and Completeness results will also be presented. Finally, the talk will outline some connections between this clausal intuitionistic logic and several familiar forms of non-monotonic reasoning. ------------------------------ Date: Thu, 19 Dec 85 14:10 EST From: Tim Finin Subject: Seminar - Robot Control with Kinesthetics (UPenn) Ph.D. Dissertation Proposal Presentation 3:00pm Friday, Dec. 20, 1985 554 Moore School University of Pennsylvania ROBOT CONTROL WITH KINESTHETICS: A REAL TIME EXPERT SYSTEM Russell L. Andersson The utility of robots is directly determined by the activities they can perform and the speed with which they can perform them. The purpose of this research is to develop means to improve the performance of robots, both in speed and especially in functionality. We maintain that the robots are already being limited by their controllers; rather than redesign the robot itself, we address the problem of how to improve the controller. Our approach is to create a sense of kinesthetics: to make more information about the robot available to the controller, and to find ways to use it. Kinesthetics requires operation in both the numeric and symbolic domains, so the system must be capable of using and interconverting both domains. We propose an expert system and a means to compile it to a form executable in real time. ------------------------------ Date: Thu, 19 Dec 85 14:10 EST From: Tim Finin Subject: Seminar - HORNLOG: Horn Clause Logic with Graph Rewriting (UPenn) Ph.D. Dissertation Proposal Presentation 3:00pm Friday, Dec. 20, 1985 554 Moore School University of Pennsylvania INVESTIGATIONS INTO HORNLOG: A HORN CLAUSE INTERPRETER BASED ON GRAPH REWRITING Stan Raatz HORNLOG, due to Jean Gallier, is a Horn clause proof procedure that can be used to interpret logic programs. The system is not resolution-based, but rather is based on a form of graph-rewriting and a linear-time algorithm for testing the unsatisfiability of propositional Horn formulae. HORNLOG applies to a class of logic programs which is a proper superset of the class of logic programs handled by PROLOG systems. In particular, negative Horn clauses used as assertions and queries consisting of disjunctions of negations of Horn clauses are allowed. This class of logic programs admits answers which are indefinite, in the sense that an answer can consist of a disjunction of substitutions. The method does not use negation by failure semantics in handling these extensions. In this proposal, using the HORNLOG procedure, we develop a theory and give examples of a type of logic programming called "general Horn clause programming. We give soundness and completeness results, show that the procedure has an immediate parallel interpretation, and argue that the method compares favorably with SLD-resolution in a parallel environment. In addition, two extensions are outlined: (1) the inclusion of term-rewriting to handle certain instances of equational programming, and (2) an uncertainty calculus for expert system applications. ------------------------------ Date: 16 Dec 1985 1005-EST From: Lydia Defilippo Subject: Course - Deduction and Computation (CMU) [Forwarded from the CMU bboard by Laws@SRI-AI.] DEDUCTION AND COMPUTATION A Spring Seminar Proposal Gerard huet Course No.: 15-850B Starting Date: 9 January 1986 Tuesdays and Thursdays from 10:00 - 11:30 am Place: 5409 Wean Hall The seminar will consist in two distinct phases. The first phase will be a standard course, and will last 8 weeks. The topics covered include basics of proof theory, such as sequent calculus, equational logic, canonical rewriting and natural deduction, and foundations of applicative programming languages, such as lambda calculus, combinators, sequential computations, types and polymorphism. The course ends with a more advanced section on the Constructions Calculus. The course is completely self-contained, there is no pre-requisite. Course notes will be available. The second phase, for the remaining 6 weeks, will be an advanced seminar on the computer automation of constructive mathematics. The participants will group themselves in teams, and attempt to develop mechanical proofs of selected theorems on the implementation of the Constructions Calculus. A team may consist in 1 to 3 participants. It is expected that terms will balance mathematical and programming talents for maximum efficiency. Each team will have access to a computer running an implementation of the Constructions Calculus. The goal of the seminar is to produce a uniform document describing the results of the experiments. ------------------------------ Date: Tue 17 Dec 85 18:48:00-CST From: AI.HASSAN@MCC.ARPA Subject: Course - Symbolic Computation (UTexas) [Forwarded from the UTexas bboard by Laws@SRI-AI.] SYMBOLIC COMPUTATION Graduate Seminar Proposal CS395-T Spring 1986 Dr. Hassan Ait-Kaci This course is intended as a bridge between Theory and Practice. It will consist of a tutorial on symbolic computation models based on functional, algebraic, and logical calculi, as well as advanced state-of-the-art in next-generation programming language research. The main focus will be on explaining how some abstract concepts derived from Logic and Universal Algebra can be used to describe symbolic computation and data structures, and how these concepts may be implemented efficiently. During the course, many existing as well as original experimental examples will be scrutinized in detail. TENTATIVE OUTLINE 1. Mathematics for Symbolic Computation a. Lambda-Calculus b. Universal Algebra c. Predicate Logic 2. Functional Computation a. Applicative Programming b. Interpreting the Lambda-Calculus c. Compiling the Lambda-Calculus (SECD Machine) d. Extensions i. Delayed-Evaluation and Streams ii. Non-Deterministic Computations iii. Binding by Unification 3. Algebraic Computation a. Term Rewriting b. Knuth-Bendix Completion Method c. Rewriting Termination d. Resolution of Equations (Narrowing, Congruence Closure) e. Equational Logic Programming 4. Logic Computation a. Horn Clause Resolution b. Compiling Horn Clause Resolution c. Extensions i. Building-In Equations ii. Building-In Inheritance iii. Non-Horn Logic Programming iv. Higher-Order Logic Programming 5. Unifying Principle: The Categorical Abstract Machine 6. Data Types a. Algebraic Abstract Data Types b. Polymorphic Types (Universal and Existential) c. Constructive Type Theory PREREQUISITE: Although meant to be self-contained, this course will offer optimal benefit to anyone acquainted with senior-level discrete mathematics and computer programming. The only *real* prerequisite is an open mind. PARTICIPATION: Students will be expected to participate actively in presentations of assigned readings, implementation of experimental programs, and in completing either a term paper or a programming project. TEXT: Most of the material covered will be taken from articles, and extensive notes written by the instructor. A (far from exhaustive) list of recommended books would be (1) Burge's "Recursive Programming Technique", (2) Henderson's "Functional Programming" (3) Campbell's (Ed.) "Implementations of Prolog", as well as anything else related to the outlined topics. ENROLLMENT: This course will be given only if at least 5 students register. Anybody interested in *registering* is encouraged to contact Hassan Ait-Kaci at 834-3354, ARPA address "Hassan@mcc", AS SOON AS POSSIBLE! ------------------------------ End of AIList Digest ******************** From uucp Mon Feb 10 02:47 EST 1986 >From SRI-AI.ARPA!LAWS Mon Feb 10 02:25:15 1986 remote from seismo Received: by seismo.CSS.GOV; Mon, 10 Feb 86 02:25:15 EST Resent-Message-Id: <8602100725.AA00770@seismo.CSS.GOV> Message-Id: <8602100725.AA00770@seismo.CSS.GOV> Mail-From: LAWS created at 20-Dec-85 09:39:39 Date: Fri 20 Dec 1985 09:34-PST From: AIList Moderator Kenneth Laws Reply-To: seismo!SRI-AI.ARPA!AIList Us-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #191 To: SRI-AI.ARPA!AIList Resent-Date: Sun 9 Feb 86 22:26:37-PST Resent-From: Ken Laws Resent-To: vtisr1!fox Status: R AIList Digest Friday, 20 Dec 1985 Volume 3 : Issue 191 Today's Topics: Queries - LISP Tools for the MicroVax II & Security Applications & Intensional and Higher-Order Logic & Experience with the TI Explorer, AI Tools - TI Superset of Common LISP & Object-Oriented Programming, Psychology - Dreams & Lateral Thinking, AI Tools - Equational Logic Programming Language ---------------------------------------------------------------------- Date: Thu, 19 Dec 85 11:41:14 GMT From: Topexprs%cs.ucl.ac.uk@cs.ucl.ac.uk Subject: LISP Tools and Environments for the MicroVax II I'd like to know about any LISP based, KBS development environments for the Microvax II. I'm looking for something that runs under microVMS, is up and running OK, more or less debugged, reasonably documented, and available now; is this too unrealistic? All useful pointers would be appreciated, also approx. prices etc. I'm used to LOOPS etc., so I'm not too keen on going to far down market. Please reply to HWB1.ARE @ CAM.PHX @ UCL-CS.ARPA Thanks, Hal Blackburn. ------------------------------ Date: Thu, 19 Dec 85 07:43 CST From: Sankar Virdhagriswaran Reply-to: Sankar Virdhagriswaran Subject: AI technology application to computer security Is anyone out there aware of applications of AI technology for computer security issues. I am aware of the pseudo object oriented models being used in developing completely secure systems. Is there more of this kind of research. I will collect the responses and post it in the net thanks in advance send replies to Araman -at hi-multics ------------------------------ Date: 19 Dec 85 10:25:00 EDT From: "CUGINI, JOHN" Reply-to: "CUGINI, JOHN" Subject: intensional and higher-order logic Anyone doing any work on using intensional or higher-order logic in knowledge representation? Any references (books, articles) would be appreciated - thanks. John Cugini ------------------------------ Date: Thu, 19 Dec 85 08:50:48 pst From: Rolf Pfeifer Subject: Experience with the TI Explorer How does the TI Explorer compare with other LISP Machines (e.g. Symbolics, LMI, or XEROX)? In particular: - availability of software - compatibility with Symbolics (effort to transport programs developed on Symbolics (or LMI) to the Explorer) - subjective experiences after having used it (heavily) for some time - support environment for AI applications development - getting to know the Explorer (e.g. how useful are their utilities they have apparently designed to support a new user?) - performance - other Comments of any sort welcome. Thanks. --Rolf Pfeifer, University of Zurich, Switzerland cernvax!unizh!pfeifer ------------------------------ Date: Tue 17 Dec 85 16:14:14-PST From: Christopher Schmidt Subject: Re: Superset of Common LISP... [Forwarded from the Stanford bboard by Laws@SRI-AI.] What TI's developing is a set of 4 4"*4" boards that implement an MIT CADR including a number of chips almost at wafer scale. The design differs from general purpose hardware only in the same way that a CADR does and doesn't really implement "Common Lisp" per se. When they say a "superset of Common Lisp" they mean MIT Lisp Machine Lisp plus an additional package that implements Common Lisp. This is rather like saying that a Shopsmith with the belt-sander attachment is a $1500 belt-sander that happens to come with a free Shopsmith. Whatever. The processor is called a "Hummingbird" and TI gives a talk on it locally every six months or so though they seem to have been by invitation. --Christopher ------------------------------ Date: Wed 18 Dec 85 20:51:44-EST From: Randy Haskins Subject: Object-Oriented Programming Re: AIList Digest V3 #189 > Flavors: (send window-1 :expose) > CORBIT: (expose window-1) > > What is seen as a message in Flavors is a function in CORBIT. So it is > not the object, but the message which is functional. [...] > > Koenraad De Smedt, DESMEDT@HNYKUN52 (bitnet) I have a hard time seeing the advantage to such a system, but then, I was brought up on object-oriented programming. The advantages of smart-objects/ dumb-messages is that functions that you write can be more generic. I have written a fair number of utilities in ZetaLisp for both ZL machines, and it's unbelievably quick and easy. One of the functions I use a lot is called GET-OBJECTS-NAMED which follows. (DEFUN GET-OBJECTS-NAMED (LST STRING &OPTIONAL (MSG :NAME)) (DELETE NIL (MAPCAR #'(LAMBDA (OBJ)(IF (STRING-SEARCH STRING (SEND OBJ MSG)) OBJ)) LST))) (The actual function I wrote is much more general, but this is the part that deals with Flavor instances pretty well.) Since :NAME is a fairly common operation for most objects to support (processes, most streams, zmacs-buffers), this function writes itself. Another example is the fact that the PRINT function will see if its argument is willing to handle a :PRINT-SELF method and will do that instead of trying to deal with the object. The same is true for DESCRIBE; I could go on for hours, but this article's already long enough. Besides, I have to get back to my ZL-style Flavors I'm implementing in MacLisp. Random ------------------------------ Date: Wed, 18 Dec 85 15:55 CDT From: Randy_Boys Subject: Response to query regarding dreams, the unconscious, and revelations In response to a posting of 29-Nov-85 (G. Joly - Subconscious Reasoning: Discovery and Invention) My background is in physiological psychology and I was a staff member at the University of Texas Health Science Center at Dallas' Cognitive Psychophysiology Laboratory for five years, doing sleep research most of the time. I have been working in the field of AI for three years now, always trying to find avenues to blend my knowledge of cognitive and computer science. I'm not sure that the following is very helpful to the progress of either field, but I was compelled to respond to Gordon and Ken's comments regarding "exceptional" insights that seemed to be associated with dream state cognition. Subconscious reasoning (if there is such a thing) is NOT the same as dream state cognition. The cortex is active during REM sleep (I'm not going to go into the REM = dream state issue) but may not be "processing" anything more than random phasic events (similarly, I'm not going into the nature or genesis of the dream state). Dreams do, however, exist and often reflect "subconscious" elements. To assume that there is some particular significance to this "window on the subconscious," as opposed to anything that may exist for "conscious" cognition, would be stretching the state of our knowledge (not that this should slow down the theorists!). In light of the fact that 2 billion people dream 4-5 times every night of their life and that only a few (and usually highly personal) dream "revelations" are cited, my response to postulations about dreams and discovery is "so what?" I do not believe that this is an area of cognition that is well enough understood to direct our inquiries into AI. Randy Boys boys@ti-eg p.s. - after 30 years of heavily funded research, science can tell us what sleep is not, but not what it is. Yes, you guessed it...the same is true of dreaming. ------------------------------ Date: Thu, 19 Dec 85 12:12:43 GMT From: gcj%qmc-ori.uucp@cs.ucl.ac.uk Subject: 42 The term ``lateral thinking'' has been in the English language for a some time. de Bono opposes lateral thinking to vertical thinking in the following sense. Vertical thinking is the logical straight- forward part of the mind, which is responsible for pattern recognition, concerned with addition and gradual modification. Lateral thinking is more concerned with making the best possible use of the information that is already available, rearranging it so that it is snapped out of the established pattern. This is a paraphrase of de Bono's words, and it would take much more space to give a really good account. The proposal that I want to suggest is that de Bono's model fits into a plan for the hopes of AI research. The vertical part of thought processes could be carried out by a machine but the lateral (creative, humorous) part could not. In mathematics, the notion exists of ``factoring out'' onto a sub- space, ie projecting down into a lower dimensional region. Perhaps a defintion of machine intelligence could be that part of the mind this left after the process of factoring out lateral subspace. Gordon Joly, gcj%qmc-ori@ucl-cs.arpa ------------------------------ Date: Fri, 20 Dec 85 9:14:41 EST From: Robert Strandh Subject: Equational Logic Programming Language EQUATIONAL LOGIC AS A PROGRAMMING LANGUAGE Michael J. O'Donnell Robert I. Strandh The University of Chicago The Johns Hopkins University Department of Computer Science Department of Electrical Engineering Ryerson Hall and Computer Science 1100 East 58th Street Barton Hall Chicago, Illinois 60637 Baltimore, Maryland 21218 odonnell@uchicago.csnet strandh@hopkins.arpa A processor for an Equational Logic Programming Language is available for distribution to Berkeley UNIX 4.2BSD VAX installations. To get a general idea of the capabilities of the interpreter, see "Programming With Equations", by Christoph M. Hoffmann and Michael J. O'Donnell, ACM ToPLAS, v. 4, no. 1 (January 1982) pp. 83-112. A user's manual is included in "Equational Logic as a Programming Language", by Michael J. O'Donnell, MIT Press, Cambridge, Massachusetts (1985). The main novelties of the interpreter are 1) strict adherence to simple semantics based on logical consequences of the given equations; 2) "lazy evaluation" (outermost evaluation) applied to all operators; 3) an implementation based on table-driven pattern matching, so that there is no run-time penalty for large sets of equations. A first experimental version was distributed in 1983. Recently, Robert Strandh has replaced the table-driven interpreter written in Pascal by compiled VAX machine code, producing an order-of-magnitude improvement in performance. Preliminary timings indicate a performance between interpreted and compiled Franz LISP. The entire interpreter system, including source files, occupies about 3.5 megabytes. ARPANET users may acquire a copy on line by executing ftp hopkins then logging on as "anonymous'' with password "anonymous", and executing the ftp commands: cd pub get equations.tar The copying process will take an hour or more. After copying the interpreter, please send a message indicating where you have installed it. Others may acquire a 1600 BPI tape copy, in "tar" format, by writing to Michael J. O'Donnell. Once the distribution file has been acquired, it should be processed by the command tar x (for the tape) tar xf equations.tar (for the tar file) Then, the instructions 2-4 in the file README should be followed. The system is in the public domain, and may be copied freely. We request notification from each site installing a copy. To defray the costs of distribution, we also request a donation of $50 to The University of Chicago for each tape, and of $5 to The Johns Hopkins University for copies taken on the ARPANet. ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Wed Jan 1 12:00:52 1986 Date: Wed, 1 Jan 86 12:00:26 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a001212; 27 Dec 85 18:41 EST Date: Fri 27 Dec 1985 15:25-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #192 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Tue, 31 Dec 85 22:15 EST AIList Digest Saturday, 28 Dec 1985 Volume 3 : Issue 192 Today's Topics: Archives - The Lisp Museum & Oxford Text Archive, Expert Systems - 3rd Jnt. BCS and ACM Symposium, Literature - Recent Articles on AI ---------------------------------------------------------------------- Date: Monday, 23 December 1985, 14:12-CST From: Oliver Gajek Subject: News Item - The Lisp Museum The LISP Museum was founded by Herbert Stoyan. It's aim is to gather each paper on LISP which was ever printed (at least all important ones): Books papers in periodicals and conference proceedings, memos manuals and adver- tisements. There's a complete list of all objects which contains over 900 entries. Everybody is invited to ask for a copy of this list. It's netto price is $5. (We hope to have the list available on a computer in the US soon.) For the appropriate amount we copy (if the copyright is open) every object in the museum. Paying in advance is a necessary condition! In addi- tion, everybody who owns some material which the museum has not available is kindly asked to send us a copy. (We appreciate original listings of old and famous LISP-programs like SIR, MLISP, MLISP2 etc. etc.) The address is: The LISP-Museum, c/o Herbert Stoyan, IMMD6, University of Erlangen, Martensstr.3, D-8520 Erlangen, Germany Oliver Gajek Linguistics Research Center University of Texas at Austin Austin, TX 78713-7247 (512)471-4166 Internet: Gajek@R20.UTEXAS.EDU uucp: ...allegra!ut-ngp!gajek ------------------------------ Date: Fri 13 Dec 85 17:38:07-CST From: Martha Morgan Subject: Pointer to Oxford Text Archive [Forwarded from the IRList Digest by Laws@SRI-AI.] Recently John Roach enquired of IRLIST, "Is there an index available of machine readable texts?" You replied, "The Oxford Text Archive has an index of its holdings..." I would like to get in touch with the Oxford Text Archive and become aware of what is contained in the index of their holdings. Can you give me an address, name, whatever? [Note: The Oxford University Computing Service Text Archive was established in 1976 as a repository of machine-readable texts to serve the research needs of scholars working with one of many different languages. It is one of the largest such collections. Texts are in many different formats and some are cleaner than others. Enquiries about texts held at Oxford or about the Archive can be sent to ARCHIVE at UK.AC.OX.VAX3, via the British Joint Academic Network. Address is Oxford Text Archive, Oxford Univ. Computing Service, 13 Banbury Rd, Oxford OX2 6NN. A modest payment and signed declara- tion form indicating that tapes will only be used for research is required. The Archive will send you forms and a list of holdings, indicating size of each item (which suggests how much tape is needed.) Be aware that many items are direct from the typesetter, without much information to help in decoding them. - Ed Fox] ------------------------------ Date: Sat, 14 Dec 85 00:11:33 cst From: "V.J. Raghavan" Subject: Abstracts - 3rd Jnt. BCS and ACM Symposium [Excerpted from the IRList Digest by Laws@SRI-AI.] Selected Abstracts from the Proceedings of the third joint BCS and ACM Symposium, Kings' College, Cambridge, 2-6 July 84 by G. Salton (Proceedings published by Cambridge Univ. Press, Editor: C.J. van Rijsbergen) BRIDGING THE GAP BETWEEN AI AND IR W.S. Cooper School of Library and Information Studies University of California Berkeley, California 94720 Information retrieval, in the broadest sense of the term, includes a conern with 'expert' or 'knowledge-based' systems and their potential future successors. It is unlikely that sophisticated systems of this sort can be developed in such a way as to use an entire natural language without the assistance of an advanced, unified theory of language and logic. The need for and probable character of such a theory are discussed. KNOWLEDGE BASED SYSTEMS VERSUS THESAURUS: AN ARCHITECTURE PROBLEM ABOUT EXPERT SYSTEMS DESIGN Mr. B. Defude Equipe Systemes intelligents de recherche d'information Laboratoire Genie Informatique (IMAG) BP 68 38402 St. Martin D'heres Cedex FRANCE The use of expert systems (ES) within information retrieval systems (IRS) seems to be an interesting way, particularly for the query process. Nevertheless we must examine what kowledge we need. We think that the thesaurus may be the kernel of which knowledge: for this, we must define it larger than in classical IRS. After some recalls about what may be the principal features of a query ES, we discuss about the relationship between thesaurus and a query expert system. The problem is to determine if the thesaurus must be integrated within the knowledge base. In fact this choice is an architecture problem of the ES. We analyze, in parallel, the effects of this choice about thesaurus representation, ES functionalities, ES architecture. The choice of an architecture depends on the goal searched: i.e. a general IR expert system able to handle a set of thesauri (independent thesaurus) or a specialized IR expert system which can be very performant but strongly tied to a specific area (integrated thesaurus). [...] ------------------------------ Date: 25 Dec 1985 23:29-CST From: leff%smu.csnet@CSNET-RELAY.ARPA Subject: Recent Articles %A Peter Friedland %A Laurence H. Kedes %T Discovering the Secrets of DNA %J CACM %V 28 %N 11 %D NOV 1985 %P 1164-1186 %K biology theory formation MOLGEN trp-operon heat shock %X describes work on an AI tool to assist scientists in theory formation They are using the Yanofsky trp-operon system as a testbed. %A Alan B. Chambers %A David C. Nagel %T Pilots of the Future: Human or Computer %J CACM %V 28 %N 11 %D NOV 1985 %P 1187-1199 %K aviation airplane %X general overview of automation techniques for aircraft piloting including AI issues %A Eric Bender %T Guru AI Environment Out %J ComputerWorld %D NOV 18, 185 %V 19 %N 46 %P 47+ %K Micro Data Base Managment System expert shell database telecommunications mainframes configuration shipping natural language %X Guru, priced at $2995, will allow users create expert systems for their applications and link those systems with database and natural language front ends. The system also integrates spread sheets, forms management, statistics, mathematics, report generators, business graphics, word processing and communications. MDBS uses GURU internally to relieve the overnight shipping manager of answering questions. 15 rules resolve 75 percent of the questions. General Electric has used the system for configuring comunications for mainframe installations. %A O. Richard Fonorow %T Users Press Icon into Commercial Service %J ComputerWorld %D NOV 18, 185 %V 19 %N 46 %P 75-92 %X tutorial on ICON, the language from Ralph Grisworld, the founder of SNOBOL. This language is recommended for AI work. %T Advertisement %J ComputerWorld %D NOV 18, 185 %V 19 %N 46 %P 138 %K WIZDOM Microcomputer Expert System %X Ad for Wizdom which is an expert system for commercial applications Software Intelligence Laboratory, Inc. Department C, 1593 Locust Ave, Bohemia, NY 11716 (516) 589-1676 %T News %J ComputerWorld %D NOV 11, 185 %V 19 %N 45 %P 12 %K Edward Feigenbaum Netherland Europe %X Edward Feigenbaum addressed the inaugural seminar for Netherlands's national research institute for knowedge systems. He predicted the main uses for AI would be speech generation, factory automation and financial applications %A David Wyland %T Software that Learns %J ComputerWorld %D NOV 11, 185 %V 19 %N 45 %P 93-104 %K Prolog Lisp %X tutorial on Prolog, Lisp and Forth, NOT machine learning %T New Products %J ComputerWorld %D NOV 11, 185 %V 19 %N 45 %P 127 %K Sourceview Micromind expert system microcomputer Macintosh %X Sourceview Software International has released Micromind Knowledge Engineering Tool for creating knowledge based expert systems on the the Macintosh. It is a rule based system. Price $495.00. %A Steven Burke %T Natural-Language Speech Recognition Product Announced %J InfoWorld %V 7 %N 47 %P 3 %K microcomputer %X Dragon Systems Inc. has developed a research prototype program that allows an IBM PC AT to recognize 2000 words spoken by a single person. It will be on the market within 18 months and cost approximately $3000. %T Full Speed Ahead for Britain's Fifth-Generation Computer %J Electronics %D DEC 9, 1985 %V 58 %N 49 %P 11 %K ICL Alice Flagsip International Computers Limited Inmos. %X The British Government agreed to provide 60 per cent of the cost for the Flagship system. This is a successor to Alice, a graph-reduction machine. One of the sixteen nodes went on line at ICL. The system will be built from Inmost transputers. %T New Products %J ComputerWorld %D DEC 9, 1985 %V 19 %N 49 %P 92 %K Expert Systems International microcomputer prolog expert system %X Expert Systems International has released version 2 of its expert systems development shell, ESP advisor, written in Prolog. The system supports 3000 rules and can invoke programs written in Prolog. %A I. Peterson %T Soliloquy for a Comptuer's Ear %J Science News %V 128 %N 23 %D DEC 7, 1985 %P 359 %K speech recognition Dragon Systems stochastic modeling IBM %X discusses the Dragon technology system, which handles 2000 words and responds in less than a second. Both the Dragon and IBM speech recognitions systems use "stochastic modeling." %A Donald F. Baxter, Jr. %T Forging Shapes Up %J Metal Progress %D DEC 1985 %P 26-33 %V 128 %N 8 %K Camel Robotics Letts Pioneer Forge Teksid Alfa Romeo Italy Forjas de Basouri Peugeot Renault France FMC General Motors crankshafts ALPID 2.0 metallurgy die manufacturing blocker %X A CAE system was developed for design and manufacture of dies being developed for the Air Force by Schultz steel. AI components are used for the automatic design of blocker and finisher dies. * An Automatic Forging Design Program, three expert systems are used sequentially to design a geometry for forgers. The first system transforms the forging geometry to design features. The second step determines parting line, forging plane, finish allowance, web thickness, rib width, draft angle, corner radii and fillet radii. The third step constructs and validates the forging geometry. * a system to design blocker die cross sections for the closed die forging of rib-web type parts. * FORMEX, to select the optimum sequence of steps in performing extrusions * Describes use of robotics at many plants to load and unload forging operations. Use of robotics can increase productivity a factor of three over human loaded systems. %A Stewart A. Denenberg %T A Service Project for an Introductory Artificial Intelligence Course: Implementing SOLO in LOGO %J SIGCSE %V 17 %N 4 %D DEC 1985 %P 8-20 %K semantic net Cognitive Psychology %X describes an AI course at State University of New York at Plattsburgh where the student implemented a system called SOLO to model human memory in LOGO. %A Jerry Lyman %T Expert Systems Tackle VLSI Testing %J Electronics %D NOV 25, 1985 %V 58 %N 47 %P 56-57 %K testability diagnosis expert systems Teradyne J941 GenRad Texas Instruments BLISS Kyushik Son Cirrus Computers %X Teradyne has expert systems for in-circut fault analysis and an expert system for diagnosing system faults in the J941 very large-scale-integeration logic designer. The diagnostic system selects appropriate in-circuit tests for each device and will deal with constraints from circuit topology and user needs. The system will then analyze test failures and determine additional actions needed after a failure to perform diagnosis. GenRad has a system to check the testability of digital IC designs. The system uses a simulator running Hitest and software by Cirrus Computers which determines apporpriate wave forms to use. Texas Instrumetns has a system for classifying switch-level faults in CMOS circuits. %A Bernard Conrad Cole %T A Pride of New CPUs Runs High-Level Languages %J Electronics %D NOV 25, 1985 %V 58 %N 47 %P 58-60] %X Lisp Machine Texas Instruments Xenologic Tantivity Associates %X Texas Instruments is developing a Compact Lisp Machine, a system designed to support Common Lisp. They expect to be sampling the chip near the end of 1985. It is fabricated in sub-2-micron CMOS and will operate at 40 Megahertz. * Xenologic has developed a two board system for prolog systems. It can perform between 200,000 and 300,000 lips. * Tantivity Associates is designing a structured language computer that can be configured to be a Lisp machine as well as a Pascal, C, Ada or Forth machine. %A Joseph J. Lazzaro %T Talking Instead of Typing %J High Technology %D JAN 1986 %V 6 %N 1 %P 58-59 %K speech recognition Kurzweil IBM Votan %A Hugh Aldersey-Williams %T Computer Eyes Turn to Food %J High Technology %D JAN 1986 %V 6 %N 1 %P 66-67 %K vision agriculture %X discusses applications of computer vision to machine harvesting or quality control in the food industry. University of Florida is developing a system to pick citrus while Michigan State is working on a system to pick strawberries. Arthur D. Little is developing a system to insure that frozen mixed vegetables have the correct proportions. %A J. Robert Lineback %T MCC: The Research Co-Op's Surprising Fast Start %J Electronics %D DEC 16, 1985 %V 58 %N 50 %P 49-51 %K Proteus Lisp truth maintenance expert systems natural language %X MCC has delivered its transfer of research from its labs to the companies that sponsored it, which is ahead of schedule. These transfer include Proteus, a Lisp based expert system using truth maintenance. MCC is working on an interactive VLSI-CAD system to be written in Lisp. The human interface department has delivered examples of experimental software for syntax analysis of knowledge-based interfaces. MCC's AI program is aimed at building a massive common-sense database consisting of 10**11 bits. %A Tobias Naegele %T AT&T Builds Fuzzy-Inference Chip %J Electronics %D DEC 16, 1985 %V 58 %N 50 %P 26-27 %K parallel real-time robotics %X AT&T has developed a chip that can perform 80,000 fuzzy inferences per second. It contains 16 rules on the chip. It is intended for applications for embedded systems for missiles and robots. %A Robert T. Gallagher %T AI to Help Mechanics at Renault to Do a Better Job %J Electronics %D DEC 16, 1985 %V 58 %N 50 %P 27 %K microcomputer expert system diagnostic %X Renault has developed an expert system to be put in automobile garages to assist in helping to diagnose such areas as automatic transmissions, on the road behavior and electrical faults. The system can be easily reconfigured to deal with different car models. %T Dornier Proposes Four High-Tech Projects to Eureka %J Electronics %D DEC 16, 1985 %V 58 %N 50 %P 17 %K fire-fighting robotics %X Dornier System Gmbh has proposed to Eureka a project to develop fire-fighting robots with artificial intelligence %A George C. Steinke %A Martin D. Schussel %T Engineering by the Book.. And On-Line %J Mechanical Engineering %D NOV 1985 %V 107 %N 11 %P 56-59 %K engineering Cognition Inc. %X Cognition Inc. of Billerica, Massachussetts will soon offer an expert system cost guide as part of its Mechanical Advantage 1000 system. Mechanical Advantage allows the user to enter the geometry of an object with symbolic parameters as well as applicable engineering design equations. Then the user can change the parameters, and have the geometry as well as the results of the equations displayed. An optimization system is also integrated into the system. %A Paul Tate %T Picking Up Speed %J Datamation %D NOV 15, 1985 %V 31 %N 22 %P 64+ %K European Computer Research Center %X A consortium of Bull of France, Britain's ICL and Gemany's Siemens is developing systems for handling large knowledge bases since they believe that the main reason for using large systems would be to hold large knowledge bases. They have developed a Prolog compiler which generates C and will include coroutining which will allow the programmer to apply constraints to variables in the language. They are working on a system to integrate logic programming and object-oriented programming. They are trying to integrate PROLOG and databases and develop AI-based front ends for packages such as spreadsheets. %A A. Knaeuper %A W. B. Rouse %T A rule-Based Model of Human Problem-Solving Behavior in Dynamic Environments %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 6 %D NOV/DEC 1985 %P 708-719 %X A rule-based model of a human plant controller has been developed at Georgia Institute of Technology. It has been compared it to human operators working on a simulated chemical production plant and find that the rule based system achieved similar performance on both stability and output as well as a 61 percent agreement with the human's actions on a case by case basis. In a related paper, they compared human operators with and without training in the fundamentals of the system, i.e. with an explanation of how the plant worked. It was found that there was no difference in performance between the two groups, even in unexpected situations where such knowledge would presumably be most helpful. %A L. A. Zadeh %T Syllogistic Reasoning in Fuzzy Logic and its Application to Usuality and Reasoning with Dispositions %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 6 %D NOV/DEC 1985 %P 754-763 %A C. P. Neuman %A V. D. Tourassis %T Inverse Dynamics Applications of Discrete Robot Models %J IEEE Transactions on Systems, Man and Cybernetics %V SMC-15 %N 6 %D NOV/DEC 1985 %P 798-803 %T TI Data Systems Reaches Goal of 100 VARs signed %J Electronic News %V 31 %N 1577 %D NOV 25, 1985 %P 52 %K Gold Hill microcomputers Texas Instruments lisp %X TI has signed a distribution with Gold Hill Computers for its Common Lisp package %A Charles Babcock %A John Gallant %T IBM Unveils Tool to Restructure VM, MVS Cobol Code %J ComputerWorld %D NOV 25, 1985 %V 19 %N 47 %P 10 %K IBM software engineering COBOL %X IBM introduced a Cobol restructuring program, COBOL/SF. The prices is $125,000 or $12,5000 per month. %A Eric Bender %T Lotus Shops for Next Technology %J ComputerWorld %D DEC 16, 1985 %V 19 %N 50 %P 1+ %K microcomputer natual language GNP Development Corporation %X GNP Development Corporation developed a Human Access Language Package which allows a user to access 1-2-3 functions through simple English commands. Lotus just purchased the firm. %T Program Has Decision-Tree Feature %J InfoWorld %D DEC 18, 1985 %V 7 %N 50 %P 59 %K Texas Instruments Arborist microcomputer decision support %X TI has introduced a version 2.0 of the Arborist Package, an AI-based decision support system for the IBM and TI PC. %J Electronic News %D DEC 9, 1985 %V 31 %N 1579 %P 8 %K Symbolics Howard I. Cannon Bruce M. Gras %X Symbolics Inc. has named Howard I. Cannon director of marketing, replacing vice-president of marketing Bruce M. Gras who has resigned. %A Alain Colmerauer %T Prolog in Ten Figures %J Communications of the ACM %D DEC 1985 %V 28 %N 12 %P 1287-1310 %X Yet Another Prolog Tutorial %A Jacques Cohen %T Describing Prolog By Its Interpretation and Compilation %J Communications of the ACM %D DEC 1985 %V 28 %N 12 %P 1311-1324 %X Yet Another Prolog Tutorial (this time emphasizing compilation and interpretation of prolog). %T How to Wreck A Nice Beach %J Science News %V 128 %D NOV 16, 1985 %P 313 %X speech recognition %X Jared Bernstein and Gay Baldwin are systematically studying the differences between spontaneous and prepared speech. They found that people have surprisingly quirky and different ways of pronouncing words. %A Ivars Peterson %T Exceptions to the Rule %J Science News %V 128 %D NOV 16, 1985 %P 314 %K natural language %X talks about research into whether there exist natural languages that are not context-free ------------------------------ End of AIList Digest ******************** From csvpi@vtcs1 Fri Dec 20 02:14:14 1985 Date: Fri, 20 Dec 85 02:13:51 est From: csvpi@vtcs1.VT To: fox@vtopus (MILLER,FRANCE,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by CSNET-RELAY.ARPA id a001238; 27 Dec 85 18:55 EST Date: Fri 27 Dec 1985 15:33-PST Reply-to: AIList%sri-ai.arpa@CSNET-RELAY US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #193 To: AIList%sri-ai.arpa@CSNET-RELAY Received: from rand-relay by vpi; Tue, 31 Dec 85 22:19 EST AIList Digest Saturday, 28 Dec 1985 Volume 3 : Issue 193 Today's Topics: Query - Common LISP and OPS5, News - Job Ads on the Network, AI Tools - Superset of Common Lisp, Games - Computer Chess (Fredkin Final), Seminar Series - AI in Design and Manufacturing (SU), Course - Logic in AI and Databases (Rutgers) ---------------------------------------------------------------------- Date: Mon, 23 Dec 85 09:16:50 -0200 From: hplabs!utah-cs!seismo!mcvax!hut.UUCP!mit@ucbvax.berkeley.edu (Markku Tamminen) Subject: Common LISP and OPS5 At the Helsinki University of Technology we have been using Franz LISP but would now like to make a switch to Common LISP, however, without undue costs. We have a VAX 11/750 under BSD 4.2 and three Altos under System 5. Further we will probably soon have an ATT 3B2, also under System 5. We would like to have Common LISP on at least two or three of these machines, so that it is important for us to find not-too-expensive variants. Does anybody have a comprehensive list of Common LISPs available under Unix? If not, and if the list is non-trivial, perhaps I could compose it from replies to this query? I have a further query about public domain versions of the OPS5 production system language. Does anybody have a version in Common LISP? If not, is such a need widely felt? Would it be possible to acquire ANY version of OPS5 through the net? (We need it quickly for use in a course beginning in January.) Markku Tamminen Helsinki University of Technology Laboratory of Information Processing Science 02150 ESPOO 15 FINLAND, Tel: 358-0-4512075 (460144) seismo!mcvax!hut!mit ------------------------------ Date: Sat, 21 Dec 85 16:29 EST From: Henry Lieberman Subject: Job ads on the network High Technology Professionals for Peace, a Cambridge, Mass. organization, is starting an electronic bulletin board for ads for jobs [non-military, of course] in the computer field. Both employers and job seekers will be able to access the system confidentially. It's not up yet, but when I receive details about how to log in, I will post them. Hopefully, there will be ways to access them from most popular computer networks. They already have an employment agency [using conventional technologies] for matching non-military employers and prospective employees, and I'd encourage anyone looking for a job to write or call them. They also seek help with their project of setting up the computer bulletin board service. ------------------------------ Date: Sat, 21 Dec 85 02:04:00 -0100 From: enea!kuling!victor@seismo.CSS.GOV (Bjorn Victor) Subject: Superset of Common Lisp In case no one else already submitted this or better: Subject: Re: Superset of Common LISP... It's probably Zetalisp they're talking about. Quoting from an article whose copy is missing origin and date, regarding the release of the Texas Explorer: "The Explorer unveiling comes shortly after TI's announcement in August of a multi-million dollar contract to develop a custom VLSI LISP processor chip for the Defense Advanced Research Project Agency (DARPA) Strategic Computing Program. The LISP microprocessor, which will be software compatible with Explorer, will perform functions requiring several hundred integrated circuits in current computer systems. The chip is being designed to provide up to ten times the processing power of today's commercial symbolic processors at substantially lower cost and physical size, which will make possible many new commercial applications of symbolic processing." --Bjorn Victor UUCP: {mcvax,seismo}!enea!kuling!victor Computing Science Dept/UPMAIL ARPA: enea!kuling!victor@SEISMO.CSS.GOV Uppsala University, PO Box 2059 S-750 02 UPPSALA, SWEDEN ------------------------------ Date: 21 December 1985 2252-EST From: Hans Berliner@A.CS.CMU.EDU Subject: Computer Chess (Fredkin Final [Forwarded from the CMU bboard by Laws@SRI-AI.] In the final round of the Fredkin Masters Invitational Tourney, Hitech beat Computer Killer Tom Martinak. Martinak had participated in the 1982 and 1984 Fredkin events in which he played a total of 5 games against the best programs at that time. His result was 4.5 out of 5 (therefore the name "computer killer"). Among his 5 points were 2 points in two tries against Cray Blitz. Final Standings: NAME SCORE PLACE RATING at start Rao 8-0 I 2400 Szmetan 6.5-1.5 II 2404 Hitech 5.5-2.5 III 2255 Leverett 5 - 3 IV 2366 Nobody else is going to get more than 3.5 points and 5th place is still undecided at this writing. This is an outstanding performance by Vivek Rao, who is the number one player in the country in the 16 and under category and number 7 in the 21 and under. He was very dominant. If I may be excused for editorializing, this is also an outstanding performance for a computer. Hitech's rating is now about 2309, which is 100 points higher than any computer has ever penetrated, and it is still climbing. ------------------------------ Date: Thu, 19 Dec 85 11:53:54 pst From: Mark Cutkosky Subject: Seminar Series - AI in Design and Manufacturing (SU) Subject: A Seminar on "A.I. in Design and Manufacturing" Time: Every thursday at 12:00 noon during the winter quarter Location: Terman? (room location to be announced) For further information contact: Mark Cutkosky, Assistant Professor, Mechanical Engineering (415) 497-8100 Susan Hansen, Administrative Assistant, SIMA (415) 497-9038 Jay M. Tenenbaum, Consulting Professor, CIS (415) 496-4704 Purpose: To explore the use of A.I. tools in problems involving the design and manufacture of mechanical and electrical components. There is a growing interest both in engineering and computer science in applying A.I. methods to engineering problems. Every week, I learn of another student who is working on an expert system or a qualitative reasoning system for some new application. A goal of this seminar is to bring us together so that we can learn what other faculty and students are doing, share ideas, and perhaps share software and programming techniques. Seminar Format: Every other week a presentation will be given by a member of the A.I. community. The intervening weeks will be devoted to informal discussions of theory and practice and to student presentations of work-in-progress. The first speaker will be Dr. Jeff Pan of Schlumberger Palo Alto Research, speaking about "A.I. Support of Fabrication Processes with Applications to Manufacturing Mechanical Devices" on Thurs, January 9. Other presentations will include: * Dr. Barbara Hayes-Roth, Stanford Knowledge Systems Laboratory: "Blackboard Systems for Mechanical Design," Feb. 6. * Dr. Robert Stults, Xerox Palo Alto Reserach Center: "Design Methodology," March 6. A more complete schedule and room location will be announced prior to the first meeting. Mark R. Cutkosky ------------------------------ Date: 26 Dec 85 13:13:05 EST From: IMIELINSKI@RED.RUTGERS.EDU Subject: Course - Logic in AI and Databases (Rutgers) Seminar in Logic in AI and Databases Tomasz Imielinski Spring 1986 This time I will talk about Logic of Knowledge and Belief and it's applications in Knowledge Representation, Incomplete and Limited Resources Reasoning and Distributed Systems (Last year seminar was related to Theoretical foundations of Logic Programming). If time allows I will also cover Nonmonotone reasoning and reasoning in presence of inconsistency. I will start from the introduction of modal logics and their semantics based on Kripke Models. This will be followed by the discussion of the Hintikka's approach to the logic of knowledge and belief (In particular the books "Knowledge and belief" and "Models for Modalities"). I will be particularly interested in predicate logic of knowledge and various difficulties and paradoxes which arise when we move from the propositional case to the predicate case. The relevant complexity results related to the decision procedures will be also discussed. Next, the applications in computer science, in particular in AI and distributed systems will be discussed. I will concentrate here again on the predicate logic of knowledge since it attracted much less attention in the literature. Such issues as "awarness", "limited resource reasoning", and "approximated reasoning" will be of particular interest here. LITERATURE: Textbook in Modal logic (like Hughes and Cresswell "Introduction to modal logic" or Chellas "Modal Logic"). Two books by J. Hintikka "Knowledge and Belief" and "Models for Modalities". These books will be put on reserve in the library. Besides various recent papers will be distributed in class possibly including the papers from the Conference in Logic of Knowledge (March 1986). PREREQUISITIES: Basic Background in logic and complexity ( 509 is enough) and AI (Intro to AI) METHOD OF EVALUATION: Homeworks and Final "Take Home" Exam TIME: I will announce the time and the place of the first meeting on the bboard on january,27. ------------------------------ End of AIList Digest ********************