From comsat@vpics1 Mon Aug 26 16:01:24 1985 Date: Mon, 26 Aug 85 16:01:13 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a002315; 8 Aug 85 22:25 EDT Date: Thu 8 Aug 1985 13:07-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #104 To: AIList@SRI-AI Received: from rand-relay by vpi; Sat, 10 Aug 85 02:35 EST AIList Digest Thursday, 8 Aug 1985 Volume 3 : Issue 104 Today's Topics: Seminars - Flexible Planning (SRI) & KADS Methodology for Knowledge Acquisition (BBN) & Parallelism in Logic Programs (SU) & Expert System Toolkit (SU) & Experiments with Belief Resolution (SRI) & Purpose-Directed Analogy (Rutgers) ---------------------------------------------------------------------- Date: Wed 31 Jul 85 16:55:51-PDT From: LANSKY@SRI-AI.ARPA Subject: Seminar - Flexible Planning (SRI) How to Plan an Action When You Don't Know What to Do: A Logic of Knowledge, Action, and Communication Leora Morgenstern New York University, SRI-AIC 11:00 AM, Monday, August 5 SRI International, Building E, Room EJ232 Most AI planners work on the assumption that they have complete knowledge of their problem domain and situation, so that formulating a plan really consists of searching through some pre-packaged list of action operators for an action sequence that achieves some desired goal. Real life planning rarely works this way, because we usually don't have enough information to map out a detailed plan of action when we start out. Instead, we initially draw up a sketchy plan and fill in details as we proceed and gain more exact information about the world. This talk will present a formalism that is expressive enough to describe this flexible planning process. The talk will consist of 5 (hopefully short) parts: 1. Motivation for a flexible logic of knowledge, action, and communication, 2. Discussion of Bob Moore's modal logic of knowledge and action, its advantages, and its limitations with respect to a robust theory of planning, 3. A move towards a syntactic theory of knowledge, and a discussion of the resulting paradoxes (esp. the Knower Paradox), 4. A solution to the Knower Paradox based on Kripke's solution to the Liar Paradox, 5. A solution to the problem of knowledge preconditions. ------------------------------ Date: 5 Aug 1985 11:09-EDT From: AHAAS at BBNG.ARPA Subject: Seminar - KADS Methodology for Knowledge Acquisition (BBN) [Forwarded from the MIT bboard by SASW@MIT-MC.] BBN-AI Seminar, 9 August 1985, 10.30 a.m. 10 Moulton St., Large Conference Room 2nd floor KADS: a structured methodology for knowledge acquisition Bob Wielinga University of Amsterdam Current Expert System technology lacks a methodology and tools which support a structured development of commercial Expert Systems. This is particularly the case for the knowledge acquisition stage in E.S. development. KADS is the result of an attempt to develop a structured methodology for knowledge acquisition for E.S., and includes some preliminary support tools. The KADS methodology is based on the following principles: 1) a decomposition of the knowledge acquisition task, 2) the use of a number of techniques for elicitation and interpretation of verbal data, 3) the formalization of verbal data in terms of epistemological models, independent of implementation details, and 4) the use of generic models for expert problem solving behaviour to guide the knowledge analysis. The KADS methodology is being implemented in a system that will support the knowledge engineer, both in performing the analysis task and in the production of documentation. In a number of case studies KADS has been used in designing and implementing expert systems. A qualitative evaluation of these studies will be presented. ------------------------------ Date: Fri, 2 Aug 85 16:33:25 pdt From: Moshe Vardi Subject: Seminar - Parallelism in Logic Programs (SU) AND/OR PARALLELISM IN LOGIC PROGRAMS by Simon Kasif Department of Computer Science University of Maryland, College Park MJH 352 Aug. 14, 1:00pm The separation of logic and control in logic programs has been shown to allow the programmer to write declara- tively lucid programs whose execution is determined by the interpreter. This appealing characteristic of logic program- ming spurred research directed towards diversifying the means for controlling the execution of logic programs. In particular, parallelism in logic programs may be exploited even when it is impossible to state a priori that two goals may be executed concurrently, but such an opportunity may be detected during the course of the execution. This talk will address the problem of AND/OR parallel- ism in logic programming. We describe a computational model for AND/OR parallel execution of logic programs. The model provides the primitives to describe and analyze parallel interpreters, emphasizing the data-flow among conjunctive goals. The effectiveness of our computational model is esta- blished through its ability to support both old and new com- munication paradigms for the parallel interpretation of logic programs. Several methods to implement AND/OR parallelism in logic programs are investigated based on notation developed in the model. The methods are shown to define a spectrum of communication schemes, ranging from the set intersection method where communication is eliminated altogether, through methods based on producers-consumers, where communi- cation is uni-directional and finally ending at a flexible bi-directional scheme introduced in the paper, called the Intelligent Channel. The primitives that comprise the model are used to syn- thesize two new parallel interpreters: Disjunctive System (DS) interpreters and Intelligent Channel Interpreters. The Intelligent Channel is a scheme we propose to constrain the combinatorial explosion of active processes, and to elim- inate the need to maintain a separate binding environment for every active OR-branch. ------------------------------ Date: Tue 6 Aug 85 16:07:30-PDT From: Carol Wright/Susie Barnes Subject: Seminar - Expert System Toolkit (SU) SIGLUNCH DATE: Friday, August 9, 1985 LOCATION: Chemistry Gazebo, betweeen Physical and Organic Chemistry TIME: 12:05 SPEAKER: Peter Jackson, Department of Artificial Intelligence, University of Edinburgh TITLE: A Flexible Toolkit for Building Expert Systems. This presentation describes the progress to date of a three-year Alvey-funded project to study, design and implement tools for building knowledge-based systems. The parties involved are Edinburgh University's Department of Artificial Intelligence and GEC's Artificial Intelligence Group at Great Baddow. The aim of the seminar is not to present finished work (the project is only six months old!), but rather to air our ideas and prejudices in the hope of attracting criticism and other kinds of feedback from the expert systems community. Our survey of current expert systems technology has led us to believe that neither shells such as EMYCIN (and derivatives) nor high-level programming languages (such as LOOPS) represent the last word in expert system building tools. The former are generally restrictive with respect to both the representation of knowledge and the specification of control, while the latter present the average programmer with a bewildering array of possibilities with little indication of how one combines different programming styles in the construction of an expert systems architecture. Thus, although there are groups of users for whom shells and AI programming languages are well-suited, we feel that there is a substantial gap in the market between the relative beginner or non-programmer, for whom the majority of commercial shells are intended, and the veteran hacker, for and by whom systems like LOOPS were developed. The alternative approach that we are currently exploring can be summed up by a number of slogans: (1) The process of choosing a logically adequate representation language and a heuristically adequate control regime and embedding these into a suitable architecture should be guided by some analysis of the task one wants one's system to perform. (2) It is worth attempting to establish a correlation between a taxonomy of expert systems tasks and representational schemes based on logical languages, with respect to both the expressiveness required by the task (e.g. modal and temporal notions, fuzziness, etc) and the control of inference (e.g. different problem solving strategies). (3) It is worth attempting to establish a similar correlation between tasks and problem solving paradigms (such as ends-means analysis, hypothesize and test, etc), with a view to helping the user decide on an architecture within which he can embed the interpretation of this chosen logical language. The problems we are currently considering include the following: (1) Is it possible to provide, along with the toolkit, an abstract architecture that can be instantiated in different ways to implement different problem solving paradigms? (2) Could one then embed different interpretations of different logics in this architecture as part of the instantiation process? (3) Could one get the behaviour associated with different problem solving paradigms from this instantiated architecture by running the logical language under different meta-level regimes? (4) Will knowledge bases created for use with one instantiation of the architecture have to be 'recompiled' for use with another instantiation? (5) How can we help a user to make the 'right' design decisions (assuming we know what these are)? We feel that this research program raises a number of very difficult issues, many of which will not be solved within the scope of the present project. Nevertheless, we also feel that practical advances in expert systems development ultimately depend upon theoretical issues of this kind being addressed, however inadequately. We still have open minds with regard to the kinds of utility that a toolkit should provide, and are always ready to talk to both the builders and users of tools in order to try and gain new insights into the problem. ------------------------------ Date: Wed 7 Aug 85 11:26:38-PDT From: LANSKY@SRI-AI.ARPA Subject: Seminar - Experiments with Belief Resolution (SRI) Experiments with Belief Resolution Kurt Konolige and Christophe Geissler SRI AI Center 11:00 AM, Monday, August 12 SRI International, Building E, Room EJ232 In recent work, Konolige developed a resolution rule for a quantified modal logic of belief. However, the rule is difficult to apply in practice, because it takes an arbitrary number of input clauses, and some instances of the rule may subsume others. In this talk we describe a solution to this problem based on a generalization of semantic attachment, that controls the growth of the search space. We have implemented the resulting version of belief resolution with Stickel's first-order connection-graph theorem prover. We present several examples of automatic reasoning about belief using this system, including a solution to the wise man problem. ------------------------------ Date: 7 Aug 85 09:37:41 EDT From: PRASAD@RUTGERS.ARPA Subject: Seminar - Purpose-Directed Analogy (Rutgers) MACHINE LEARNING SEMINAR Title: Purpose-Directed Analogy** Speaker: Smadar Kedar-Cabelli Date: Monday, August 12, 11:00 AM Place: Hill Center, room 423 Recent artificial intelligence models of analogical reasoning are based on mapping some underlying causal network of relations between analogous situations. However, causal relations relevant for the purpose of one analogy may be irrelevant for another. In this talk, I will introduce a technique which uses an explicit representation of the purpose of the analogy to automatically create the relevant causal network. I will illustrate the technique with two case studies in which concepts of everyday artifacts are learned by analogy. ** This is a dry-run for a talk being presented at the Cognitive Science Society Conference in Irvine, CA, August 15-17. ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Fri Aug 9 05:30:32 1985 Date: Fri, 9 Aug 85 05:30:24 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: RO Received: from sri-ai.arpa by csnet-relay.arpa id a001683; 8 Aug 85 20:49 EDT Date: Thu 8 Aug 1985 13:15-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #105 To: AIList@SRI-AI Received: from rand-relay by vpi; Fri, 9 Aug 85 05:17 EST AIList Digest Friday, 9 Aug 1985 Volume 3 : Issue 105 Today's Topics: Reports - Semantic Automata & AI Newsletter from TI, Call for Papers - Expert Systems for Engineering, Conferences - AI in Engineering Conference & Aerospace Applications of AI & Office Information Systems ---------------------------------------------------------------------- Date: Wed 31 Jul 85 16:53:17-PDT From: Emma Pease Subject: Report - Semantic Automata [Excerpted from the CSLI Newsletter by Laws@SRI-AI.] NEW CSLI REPORT Report No. CSLI-85-27, ``Semantic Automata'' by Johan van Benthem, 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: Wed 7 Aug 85 13:16:35-PDT From: Ken Laws Subject: AI Newsletter from TI IEEE Computer reports that: Texas Instruments has launched the Artificial Intelligence Newsletter to communicate the potentials, limitations, and progress of AI to nonspecialists. First issue presents the menu-based approach to natural languages and AI briefs. Available as a public service from TI Data Systems Group, PO Box 2909, MS 2222, Austin, TX 78769; (512) 250-6314. ------------------------------ Date: Wed, 7 Aug 85 16:23:17 EDT From: John Kastner Subject: Call for Papers - Expert Systems for Engineering CALL FOR PAPERS IEEE COMPUTER MAGAZINE SPECIAL ISSUE ON EXPERT SYSTEMS FOR ENGINEERING APPLICATIONS Contributions are hereby solicited for this special issue. To be considered, 5 copies of a manuscript must be sent, no later than October 1, 1985, addressed to: Se June Hong, Guest Editor IEEE Computer Magazine Special Issue IBM Thomas J. Watson Research Center 31-206 P.O. Box 218 Yorktown Heights, NY 10598 Any paper that is postmarked after the date will be returned to the author, unprocessed. Further information may be obtained by contacting the guest editor. Se June Hong ARPAnet: HONG.YKTVMX.IBM-SJ@CSnet-Relay 914/945-2265 ------------------------------ Date: Sunday, 4 August 1985 20:55:27 EDT From: Duvvuru.Sriram@cmu-ri-cive.arpa Subject: Two slots available in AI in Engg. Conf. There are a couple of slots open in the TOOLS session of the First International Conference on AI in Engineering, to be held in UK next April. If your company's framework was used for an engineering application, here is a good chance to present your product. Interested parties should send mail to me at sriram@cmu-ri-cive.arpa. Sriram ------------------------------ Date: 2 Aug 85 13:28 PDT From: halko.pasa@Xerox.ARPA Subject: Conference : Aerospace Applications of AI ACM/SIGART (Dayton, Ohio, Chapter) will hold an AI Conference Sept 16 - 19 (Registration Sept 16, Sessions 17-19) 1985. Registration fee $225. Address is AAAIC '85, Box 31250, Dayton, Ohio, 45431-0250. Phone 513-426-8530 While the name says "aerospace", this is because of the proximity of the big air base there and because one or two of the sessions deals with avionics. Most of the sessions are applicable to AI in general: Avionics, manufacturing, maintenance, decision support systems, expert system building tools, programming languages, man-machine interfaces, and new architectures. World famous speakers are coming, such as Dr. M. Stefik, XEROX PARC; Dr. Earl Sacerdoti, Teknowledge, Dr. M. Fox, CMU, Prof. D. Michie, Univ of Edinburgh, and many more. Attendance is limited to 700 people. Direct inquiries by mail or phone to the address above, or contact Gerald Matthews, NCR R&D WHQ-5E, Dayton, Ohio, 45479, Phone 513-445-6054. ------------------------------ Date: Thu, 8 Aug 85 02:48 EDT From: Carl Hewitt Subject: Call for papers: OIS-86 ****************************************************** * CALL FOR PAPERS * * * * THIRD ACM CONFERENCE ON * * OFFICE INFORMATION SYSTEMS: OIS-86 * * * * October 6-8, 1986 * * Biltmore Plaza Hotel * * Providence, RI * ****************************************************** General Chair: Carl Hewitt, Program Chair: Stanley Zdonik, MIT Brown University Treasurer: Gerald Barber, Local Arrangements: Andrea Skarra, Gold Hill Computers Brown University An interdisciplinary conference on issues relating to office information systems sponsored by ACM/SIGOA in cooperation with Brown University and the MIT Artificial Intelligence Laboratory. Submissions from the following fields are solicited: Anthropology Artificial Intelligence Cognitive Science Computer Science Economics Management Science Psychology Sociology Topics appropriate for this conference include (but are not restricted to) the following: Technologies including Display, Voice, Telecommunications, Print, etc. Human Interfaces Deployment and Evaluation System Design and Construction Goals and Values Knowledge Bases and Reasoning Distributed Services and Applications Indicators and Models Needs and Organizational Factors Impact of Computer Integrated Manufacturing Unpublished papers of up to 5000 words (20 double-spaced pages) are sought. The first page of each paper must include the following information: title, the author's name, affiliations, complete mailing address, telephone number and electronic mail address where applicable, a maximum 150-words abstract of the paper, and up to five keywords (important for the correct classification of the paper). If there are multiple authors, please indicate who will present the paper at OIS-86 if the paper is accepted. Proceeedings will be distributed at the conference and will later be available from ACM. Selected papers will be published in the ACM Transactions on Office Information Systems. Please send eight (8) copies of the paper to: Prof. Stan Zdonick OIS-86 Program Chair Computer Science Department Brown University P.O. Box 1910 Providence, RI 02912 DIRECT INQUIRIES TO: Rita Desormeau (401) 863-3302 ****************************************************************************** IMPORTANT DATES Deadline for Paper Submission: February 1, 1986 Notification of Acceptance: April 30, 1986 Deadline for Final Camera-Ready Copy: July 1, 1986 Conference Dates: October 6-8, 1986 ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Mon Aug 26 16:00:49 1985 Date: Mon, 26 Aug 85 16:00:34 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a000975; 9 Aug 85 17:38 EDT Date: Fri 9 Aug 1985 13:14-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #106 To: AIList@SRI-AI Received: from rand-relay by vpi; Sat, 10 Aug 85 03:28 EST AIList Digest Saturday, 10 Aug 1985 Volume 3 : Issue 106 Today's Topics: Journal - Machine Learning, Expert Systems - Knowledge-Based Spelling Corrector & Expert System Definition ---------------------------------------------------------------------- Date: Fri, 9 Aug 85 01:22:57 pdt From: gluck@SU-PSYCH (Mark Gluck) Subject: A new journal: MACHINE LEARNING Machine Learning will publish papers on the processes through which intelligent systems improve their performance over time, and will cover ares such as: concept acquisition, strategy learning, language development, reasoning by analogy, and scientific discovery. Machine Learning will be published four times a year. Individual Rate: $35.00, Institution Rate: $78.00. Executive Editor: Pat Langley Editors: Jaime Carbonell, Ryszard Michalski, Tom Mitchell Orders to: Kluwer Academic Publishers, 190 Old Derby Street, Hingham, MA 02043 or call 617-749-5262. First Issue: January, 1986 ------------------------------ Date: Wed 7 Aug 85 14:41:38-PDT From: Ken Laws Subject: Knowledge-Based Spelling Corrector Those who objected to Bob Amsler's suggestion that all spelling correctors are "knowledge-based" may be happier with Dr. Dave Fawthrop's Expert Speller. There's a 2-page article on it in Vol. 1, No. 1, of Expert Systems User, April 1985. (The article is rather superficial and contains some editing blunders, so interested readers should probably track down the true story in "The Rules of Spelling Error" and "An Intelligent Spelling Error Corrector" by E.J. Yannakoudakis and D. Fawthrop, Int. J. of Information Processing and Management, 1983.) The spelling program uses a bit-vector hash table and two slower dictionary structures to detect misspelled words, then invokes a rule base of about 3,000 rules to predict the correct spelling. The system contains some knowledge of parts of speech, but bases most of its guesses on commonly- occurring transformations of letter patterns. I assume that suggested corrections are looked up in the dictionaries before being offered to the user (although I can imagine wanting to see all its hypotheses). New words can be provided (or confirmed) during a session, so that the program adapts itself to the specific document during the editing session and to the user across sessions -- AI learning in the best tradition. The probabilities of different transformations were adaptively adjusted (at a cost of 250,000 pounds in computer time) for the best correction performance, and Dr. Fawthrop makes the point that the system now knows more than he does about correction probabilities. (The system cannot "explain its reasoning", since no human is able to "understand" the true explanation.) The program is written in Fortran77, and runs mainly on Unix systems. The article didn't supply an address. -- Ken Laws ------------------------------ Date: 7 Aug 85 15:11 PDT From: Ghenis.pasa@Xerox.ARPA Subject: Re: Expert System definition I think it was Roger Schank that said "If it can't learn, it isn't AI". In a nutshell, I think the main thing that distinguishes a true ES from the hypothetical FORTRAN program mentioned earlier is that knowledge isn't hard-coded and can be added at run time, becoming immediately useful both for inferences and explanations. This highlights the difference in architectures: separation of inference engine and knowledge base (knowledge=rules+facts) should be considered an essential part of the definition of ES or KBS, so add "DYNAMIC KNOWLEDGE BASE" to "perform tasks requiring expertise" and "ability to explain". By the way, the ability to gracefully expose and resolve knowledge base inconsistencies is also highly desirable in any system with a dynamic KB, hopefully with a precedence mechanism and/or weighting and/or consulting a human being when all else fails. ------------------------------ Date: 7 Aug 85 15:12 PDT From: masinter.pa@Xerox.ARPA Subject: Expert System Definition I believe that the term "expert system" is best thought of as a process, rather than a thing; that is, it is a way of building programs rather than something that a program can be. In this light, most of the thorny issues that arise in trying to decide whether a program is or is not an "expert system" become much clearer. The "expert system" programming methodolgy has some simple characteristics: * An expert is involved in the construction of the program. * The knowledge of the expert is couched within a representation framework in which the objects of the domain and the relations between them are explicitly represented. * The expert is involved in refinement of the program after its initial construction. To answer the question "is this FORTRAN program an expert system", you have to know "how did it get built?" and "how would you add more knowledge to it?" The various AI languages and systems make writing expert systems much easier; that's their advantage. However, their use is not criterial: writing it in OPS-5 doesn't make it an expert system, and not all expert systems are written using "expert system tools." Rule-based programming is one mechanism for encoding knowledge, but its use isn't criterial either: there are rule-based programs that are trivial (the microprocessors that control traffic lights, for example), and there are "expert systems" that do not use a rule-based paradigm for performing inference. Generally, experts aren't programmers, and their knowledge of their field isn't explicit. The process of producing domain-descriptions and rules usually involves running the system against test cases and debugging the results. "Explaination facilities" are very helpful in that debugging process, in localizing where the procedure has gone wrong. It seems to be an important part of the development tools for the expert-system methodology, even if it plays no part in final application. ------------------------------ Date: 07 Aug 85 18:01:34 PDT (Wed) From: Sanjai Narain Subject: What is an expert system Wyland is absolutely right in encouraging the asking of simple questions. I think many more simple questions need to be asked in AI. According to him an expert system is a program which can not only do reasoning but also explanation. I believe this definition is a special case of a more general one: Given some fixed amount of knowledge think of the various ways in which an (intelligent) human could use it. If a program possessing the same knowledge could also use it in similar ways then it could be said to behave intelligently. For example, when a human has medical knowledge, his intelligence allows him to do diagnosis, to explain his diagnosis as well as to determine whether a new piece of knowledge conflicts with something he already knows. So, if a program which has medical knowledge can also do diagnosis, explanation and integrity checking, it could reasonably be regarded as intelligent. Conversely, even though common word processing software allegedly "knows" about letters, words, sentences, lines, paragraphs it could hardly be expected to determine whether a sentence is well formed, or whether the numbering scheme for sections is consistent throughout the text. Such functions are normally associated with a competent copy editor. It is quite hard to obtain flexibility of use of knowledge in procedural languages. So, most software written in such languages can only be used in a rigidly defined set of ways, and so is not considered intelligent. Not surprisingly then, a central concern of AI is developing maximally flexible representations of knowledge so it can be used in a great diversity of ways. -- Sanjai Narain ------------------------------ Date: 8 Aug 1985 07:50-PDT From: JWOLFE@USC-ECLB Subject: KNOWLEDGE SYSTEM DEFINITION I read your comments on a definition of expert systems. I think the views you express are useful and sorely needed. If AI is to succeed, then it is necessary to precisely define the terms used within the discipline. However, I do have one comment. The discussion I read differentiated between a "true" expert system and a simple FORTRAN program with a sequence of IF...THEN statements by asserting that the expert system could explain its conclusions in terms of the rules used to derive those conclusions. Furthermore, there was a qualification that the expert system merely had to have the potential of explaining itself and that it was not necessary that the explanation mechanism be implemented. I submit that the distinction is artificial in that the FORTRAN program could easily implement an explanation mechanism via a stack. In my opinion, the definition of expert system should require that the explanation mechanism be implemented or drop the requirement altogether. Perhaps another distinction would be the degree of separation between the knowledge specific data and the reasoning mechanism. This would exclude programs in which the knowledge was totally embedded in the program code. I don't believe the distinction is wholy satisfactory since all but the most trivial systems have some knowledge embedded in the program in order to gain performance. The line at which a system is or is not an expert system becomes fuzzy. Thank you again for sharing your views with the community. I hope to hear from you. Jim Wolfe JWOLFE@USC-ECLB.ARPA (usual disclaimers apply) ------------------------------ Date: Thu 8 Aug 85 09:53:53-PDT From: WYLAND@SRI-KL.ARPA Subject: Expert System Definition Thank you for your comments. I should have been more precise when I said that it was not necessary to implement the explanation mechanism, since I used it as the essential ingredient of the definition of an expert system. I agree with you that for a program to be an expert system, the explanation mechanism is required. There is a marginal case of an expert system where the explanation mechanism has been implemented but is removed or disabled in the application environment, as in a real-time pilot advisor or process control program. However, the clearest definition of an expert system would allow only systems that had the explanation mechanism implemented, but not necessarily used (i.e., it could be disabled). Otherwise, the definition is void since "any" rule based system has the "potential" for the addition of an explanation mechanism - i.e., "....the FORTRAN program could easily implement an explanation mechanism via a stack." I like the definition of an expert system as a "decision system which can explain its decisions" because it is functional and objective. Using this definition, other characteristics associated with expert systems - such as the degree of separation between the knowledge data and the reasoning mechanism and/or the amount of knowledge embedded in the code - become design topics on how to implement a system that meets the definition rather than part of the definition itself. This is good because - as you pointed out - topics like "separation between the knowledge base and the reasoning mechanism" are a matter of degree and are subjective in that design/artistic judgement is involved in their evaluation. Thank you again for helping to further clarify the discussion. I hope we can keep in touch. Dave Wyland WYLAND@SRI-KL ------------------------------ Date: Friday, 9 August 1985 03:39:57 EDT From: Duvvuru.Sriram@cmu-ri-cive.arpa Subject: Discussion on Expert Systems Definition Although the ability to explain things is one of the characteristics of an Expert, only classification-type expert systems (ES) have this feature. These systems explain their actions by providing the rules that have been used or being used in that context. However, a majority of ES (or the so-called ES), such as R1, in the market do not have this feature. The main difference (with conventional languages) I see is the ease of programming with rules. Try implementing the code provided by Wyland in a conventional programming language. It may take you at least 1 to 2 hrs, while one could do the same in a rule-based framework in 10-15 minutes, assuming familiarity with the tool. The ES frameworks provide a neat programming methodology for incorporating heuristic knowledge. Also, the completeness, uniqueness, and proper sequencing criteria, required of many conventional languages can be relaxed in an ES envir- onment. Any Comments ==> AILIST Sriram ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Mon Aug 26 15:55:35 1985 Date: Mon, 26 Aug 85 15:55:27 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a010113; 11 Aug 85 18:03 EDT Date: Sun 11 Aug 1985 14:13-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #107 To: AIList@SRI-AI Received: from rand-relay by vpi; Mon, 12 Aug 85 03:26 EST AIList Digest Sunday, 11 Aug 1985 Volume 3 : Issue 107 Today's Topics: Seminar - Expert System for Statistical Application (SU) & Prolog (Rand) & The PRISM Expert System (IBM-SJ) & Parallelism in Logic Programs (IBM-SJ) & Computer Music Expert System (CMU), Conference - Foundations of AI ---------------------------------------------------------------------- Date: Thu, 8 Aug 85 22:54:01 pdt From: naomi@playfair (naomi altman) Subject: Seminar - Expert System for Statistical Application (SU) Laboratory for Computational Statistics Seminar 3:15pm, Friday Aug 9, 1985 in Sequoia 114 AN EXPERT SYSTEM OF STATISTICAL APPLICATION Knut M. Wittkowski University of Tubingen, Department of Medical Biometry Most structural information on statistical data (number and hierarchy of factors, sampling strategy, scale types) are neglected by common statistical data base management systems. The wealth of methods currently available in modern statistical program packages, consequently, often leads to erroneous applications of statistical methods. It is demonstrated, how an expert system can facilitate the use of statistical analysis systems by means of intelligent dialogue techniques based on knowledge of structural information and help to avoid erroneous applications of statistical (graphical or analytical) methods. ------------------------------ Date: 09 Aug 85 10:14:08 PDT (Fri) From: Sanjai Narain Subject: Seminar - Prolog (Rand) THE EXPRESSIVE POWER OF PROLOG Peter Schmitt IBM, Heidelberg, West Germany 2:00 p.m. Tuesday, August 13, 1985 Rand Corporation, Room 2760 This talk is concerned with the foundations of logic programming. In particular, completeness results and insufficiencies of PROLOG are discussed including questions of search strategy, occur check and negation. ------------------------------ Date: Fri, 9 Aug 85 10:47:14 PDT From: IBM San Jose Research Laboratory Calendar Reply-to: IBM-SJ Calendar Subject: Seminar - The PRISM Expert System (IBM-SJ) [Excerpted from the IBM-SJ Calendar by Laws@SRI-AI.] IBM San Jose Research Lab 5600 Cottle Road San Jose, CA 95193 Thur., Aug. 15 Computer Science Seminar 10:00 A.M. PRISM - AN EXPERT SYSTEM Auditorium While the expert system has been developed as a knowledge acquisition and delivery vehicle by the AI researchers, it has evolved to be a practical software development productivity tool. PRISM is an expert system prototype developed at the Palo Alto Scientific Center and has been available for application development for more than a year to internal users and university study partners. Recently, IBM announced its first expert system product, Expert System Environment/VM, based on PRISM. This talk will begin with an introduction to the expert system technology: its basic architecture, knowledge representation and inferencing, the interrelationship among the application domain expert, the knowledge engineer, and the client. The difference between the traditional application programming and the expert system approach will be emphasized. The second half of the talk will describe the product and some projects and applications using the PRISM technology. F. C. Tung, IBM Palo Alto Scientific Center Host: K. Wong ------------------------------ Date: Fri, 9 Aug 85 10:47:14 PDT From: IBM San Jose Research Laboratory Calendar Reply-to: IBM-SJ Calendar Subject: Seminar - Parallelism in Logic Programs (IBM-SJ) [Excerpted from the IBM-SJ Calendar by Laws@SRI-AI.] IBM San Jose Research Lab 5600 Cottle Road San Jose, CA 95193 Fri., Aug. 16 Computer Science Seminar 2:00 P.M. PARALLELISM IN LOGIC PROGRAMS Aud. A The separation of logic and control in logic programs has been shown to allow the programmer to write declaratively lucid programs whose execution is determined by the interpreter. This appealing characteristic of logic programming spurred research directed towards diversifying the means for controlling the execution of logic programs. In particular, parallelism in logic programs may be exploited even when it is impossible to state a priori that two goals may be executed concurrently, but such an opportunity may be detected during the course of the execution. This talk will address the problem of and/or parallelism in logic programming. We describe a computational model for and/or parallel execution of logic programs. The model provides the primitives to describe and analyze parallel interpreters, emphasizing the data-flow among conjunctive goals. The effectiveness of our computational model is established through its ability to support both old and new communication paradigms for the parallel interpretation of logic programs. Prof. S. Kasif, Department of Computer Science, University of Maryland, College Park Host: P. Lucas ------------------------------ Date: 7 August 1985 1700-EDT From: Roger Dannenberg@CMU-CS-A Subject: Seminar - Computer Music Expert System (CMU) [Forwarded from the CMU bboard by Laws@SRI-AI.] Marilyn Taft Thomas (Music Department) and I will each present a short talk on Monday, August 12, from 3:30PM to 4:30PM in WeH 4623. Dr. Thomas's talk is: "Vivace: A Rule-Based AI System for Composition". Vivace composes 4-part chorales in the style of Bach. Sound examples of Vivace compositions will be performed. My talk is "Real-Time Computer Accompaniment of Keyboard Performance" and is based on a paper co-authored with Joshua Bloch. The talk will cover the application of dynamic programming to on-line pattern matching of polyphonic music, and heuristics for musical accompaniment. A video-tape of our system will be shown. Both talks will be presented in a few weeks at the 1985 International Computer Music Conference. We hope to receive constructive criticism on our presentations as well as to share our latest results. ------------------------------ Date: Fri 9 Aug 85 13:39:33-PDT From: Ken Laws Subject: Conference - Foundations of AI >From CACM, August 1985: The AAAI and the Computing Research Laboratory, New Mexico State University, are sponsoring a Workshop on the Foundations of AI, February 6-8, 1986, in Las Cruces, NM. Papers dealing with the following three topics are sought: relationships between foundations and working programs; relationships between AI and other disciplines; and philosophical, logical, and theoretical foundations of AI. Three copies of a paper (maximum 2000 words) should be submitted by September 1 to Derek Partridge, Computing Research Laboratory, NMSU, Las Cruces, NM 88003. Authors are to be notified by November 1. ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Mon Aug 26 15:45:52 1985 Date: Mon, 26 Aug 85 15:45:43 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a003763; 13 Aug 85 1:39 EDT Date: Mon 12 Aug 1985 21:51-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #108 To: AIList@SRI-AI Received: from rand-relay by vpi; Tue, 13 Aug 85 21:24 EST AIList Digest Tuesday, 13 Aug 1985 Volume 3 : Issue 108 Today's Topics: Conference - Cognitive Science Society ---------------------------------------------------------------------- Date: 12 Aug 1985 1811-PDT From: GRANGER%UCI-20A@UCI-ICSA Subject: 7th Annual Conference of the Cognitive Science Society, 15-17 Aug. Though it's the last minute, I'm sending a copy of the program for the upcoming seventh annual conference of the Cognitive Science Society. The conference runs 15-17 August (immediately before ijcai), and contains much that may be of interest to the Artificial Intelligentsia. Anyone who is interested in coming down for this conference, can show up and register at the door; dorm rooms are available on the campus, and hotel rooms are available at the Sheraton Hotel in Newport Beach. The conference is being held on the campus of the University of California, Irvine, which is about 40 miles south of Los Angeles, adjoining Newport Beach. There are currently about 400 preregistered participants. -Rick Granger (Granger@UCI) ======================================================================== Thursday 15 August 1985 7:30am-8:45am Breakfast [Mesa Court Cafeteria] 8:45am-10:15am Invited Address Shimon Ullman, Massachusetts Institute of Technology [Fine Arts Village Theater] 10:15am-10:30am Break 10:30am-12:10pm Paper Session I [Fine Arts Village Theater] 10:30am-10:55am Symmetry Detection and the Perceived Orientation of Simple Plane Polygons Paul Kube, University of California, Berkeley 10:55am-11:20am Variations on Parts and Wholes: Information Precedence vs. Global Precedence Marc M. Sebrechts, Wesleyan University John J. Fragala, Wesleyan University 11:20am-11:45am The Neural Locus of Mental Image Generation: Converging Evidence from Brain-Damaged and Normal Subjects Martha J. Farah, Carnegie-Mellon University 11:45am-12:10pm A Developmental Neural Model of Word Perception Richard M. Golden, Brown University 10:30am-12:10pm Paper Session II [Concert Hall] 10:30am-10:55am A Computer Model of the Neural Substrates of Classical Conditioning in the Aplysia Mark A. Gluck, Stanford University Richard F. Thompson, Stanford University 10:55am-11:20am Structural Learning in Connectionist Systems Andrew G. Barto, University of Massachusetts, Amherst Charles W. Anderson, University of Massachusetts, Amherst 11:20am-11:45am The Learning of World Models by Connectionist Networks Richard S. Sutton, GTE Laboratories Brian Pinette, University of Massachusetts, Amherst 11:45am-12:10pm Learning Salience Among Features Through Contingency in the CEL Framework Richard H. Granger, University of California, Irvine Jeffrey C. Schlimmer, University of California, Irvine 12:10pm-2:00pm Lunch Break [Mesa Court] 2:00pm-4:45pm Invited Symposium I Attention in Early Visual Processing [Fine Arts Village Theater] Anne Treisman, University of British Columbia Michael Posner, University of Oregon David LaBerge, University of California, Irvine Francis Crick, The Salk Institute Daniel Kahneman, University of British Columbia Shimon Ullman, Massachusetts Institute of Technology 4:45pm-7:00pm Dinner Break [Mesa Court] 7:00pm-11:00pm Poster Session and Reception [University Club] (List of poster presentations at end of program) Friday 16 August 1985 8:45am-10:15am Invited Address: Allen Newell, Carnegie-Mellon University [Fine Arts Village Theater] 10:15am-10:30am Break 10:30am-12:10pm Paper Session III [Fine Arts Village Theater] 10:30am-11:20am Bounded Irrationality: The Psychology of Incoherence (paper not published in the proceedings) Daniel Kahneman, University of British Columbia 11:20am-11:45am Component Models of Physical Systems Allan Collins, Bolt Beranek and Newman 11:45am-12:10pm Temporal Notation and Causal Terminology Yoav Shoham, Yale University Thomas Dean, Yale University 10:30am-12:10pm Paper Session IV [Concert Hall] 10:30am-10:55am Story Telling and Generalization Michael Lebowitz, Columbia University 10:55am-11:20am MULTIPAR: A Robust Entity-Oriented Parser Jill Fain, Carnegie-Mellon University Jaime G. Carbonell, Carnegie-Mellon University Philip J. Hayes, Carnegie-Mellon University Steven N. Minton, Carnegie-Mellon University 11:20am-11:45am Towards a Computational Theory of Human Daydreaming Erik T. Mueller, University of California, Los Angeles Michael G. Dyer, University of California, Los Angeles 11:45am-12:10pm Integrating Marker-Passing and Problem Solving James A. Hendler, Brown University 12:10pm-2:00pm Lunch Break [Mesa Court] 2:00pm-4:45pm Invited Symposium II Integrated Empirical Models of Learning and Memory [Fine Arts Village Theater] Paul Rosenbloom, Stanford University Gary Lynch, University of California, Irvine Pat Langley, University of California, Irvine Jaime Carbonell, Carnegie-Mellon University David Rumelhart, University of California, San Diego 4:45pm-5:30pm Reception [University Club] 5:30pm-7:30pm Banquet [University Club] 7:30pm-9:00pm Invited Address Endel Tulving, University of Toronto [University Club] 9:00pm-12:00mid Reception [University Club] Saturday 17 August 1985 8:45am-10:15am Invited Address: Roger Schank, Yale University [Fine Arts Village Theater] 10:15am-10:30am Break 10:30am-12:10pm Paper Session V [Fine Arts Village Theater] 10:30am-10:55am The Evolution of Knowledge Representations with Increasing Expertise in Using Systems Dana S. Kay, Yale University John B. Black, Teachers College, Columbia University 10:55am-11:20am Purpose-Directed Analogy Smadar Kedar-Cabelli, Rutgers University 11:20am-11:45am Learning Concrete Strategies Through Interaction R.W. Lawler, GTE Laboratories Oliver G. Selfridge, GTE Laboratories 11:45am-12:10pm Failure-Driven Acquisition of Figurative Phrases by Second Language Speakers Uri Zernik, University of California, Los Angeles Michael G. Dyer, University of California, Los Angeles 10:30am-12:10pm Paper Session VI [Concert Hall] 10:30am-10:55am Two Kinds of Feature? A Test of Two Theories of Typicality Effects in Natural Language Categories Robin A. Barr, Ball State University Leslie J. Caplan, Ball State University 10:55am-11:20am Empirical Evidence for a Global Workspace Theory of Voluntary Control Bernard J. Baars, University of California, San Francisco 11:20am-11:45am Connectionist Parsing Garrison W. Cottrell, University of Rochester 11:45am-12:10pm A Rule-Based Connectionist Parsing System Bart Selman, University of Toronto Graeme Hirst, University of Toronto 12:10pm-2:00pm Lunch Break [Mesa Court] 1:00pm-2:00pm Business Meeting of the Society [Mesa Court] 2:00pm-4:45pm Invited Symposium III Syntactic Language Processing [Fine Arts Village Theater] Robert Berwick, Massachusetts Institute of Technology Lyn Frazier, University of Massachusetts, Amherst Howard Kurtzman, University of California, Irvine Eric Wehrli, University of California, Los Angeles Ken Wexler, University of California, Irvine 4:45pm-5:30pm Farewell Reception [Mesa Court] Poster Presentations 7:00pm-11:00pm Thursday, 15 August [University Club] Adaptive Planning: Refitting Old Plans to New Situations, Richard Alterman, University of California, Berkeley Memory Representation and Retrieval for Editorial Comprehension, Sergio J. Alvarado, Michael G. Dyer, and Margot Flowers, University of California, Los Angeles Analogy Recognition and Comprehension in Editorials, Stephanie E. August and Michael G. Dyer, University of California, Los Angeles Investigations of Information Utilization during Fixations in Reading, Harry E. Blanchard, University of Illinois at Urbana-Champaign Two Endorsement-based Approaches to Reasoning About Uncertainty, Paul R. Cohen, University of Massachusetts, Amherst Teleology + Bugs = Explanations, Gregg C. Collins, Yale University A Model for Understanding the Points of Stories, Marcy Dorfman, University of Illinois at Urbana-Champaign A Framework for Concept Formation, J. Daniel Easterlin and Pat Langley, University of California, Irvine The Problem of Existence, Kenneth D. Forbus, University of Illinois at Urbana-Champaign Cross-Mapped Analogies: Pitting Systematiciy Aganist Spurious Similarity, Dedre Gentner, University of Illinois at Urbana-Champaign; Cecile Toupin, University of California, Berkeley Information, Uncertainty, and the Utility of Categories, Mark A. Gluck, Stanford University; James E. Corter, Columbia University A Model of Question Answering, Arthur C. Graesser, Memphis State University; George Vamos, University of Southern California; David Koizumi, C. Scott Elofson, California State University, Fullerton The Time Course of Anaphora Resolution, Raymonde Guindon, Microelectronics and Computer Technology Corporation Using a Computational Model of Language Acquisition to Address Questions in Linguistic Inquiry, Jane Hill, Smith College A Model of Acquiring Problem Solving Expertise, Dennis Kibler and Rogers P. Hall, University of California, Irvine Creating and Comprehending Arguments, Stuart M. McGuigan, Yale University; John B. Black, Teachers College, Columbia University Levels of Goal Direction and the Causes of Learning, Dale M. McNulty, University of California, Irvine Connectionist Learning in Real Time: Sutton-Barto Adaptive Element and Classical Conditioning of the Nictitating Membrane Response, J.W. Moore, J.E. Desmond, N.E. Berthier, D.E.J. Blazis, R.S. Sutton, A.G. Barto, University of Massachusetts, Amherst Explanation and Generalization Based Memory, Michael J. Pazzani, University of California, Los Angeles and The Aerospace Corporation Bayesian Networks: A Model of Self-Activated Memory for Evidential Reasoning, Judea Pearl, University of California, Los Angeles Expert Variance: Differences in Solving a Dynamic Engineering Problem, Michael Prietula and Frank Marchak, Dartmouth College Machine Understanding and Data Abstraction in Searle's Chinese Room, William J. Rapaport, University of Buffalo Toward a Unified Model of Deception, Donald D. Rose, University of California, Irvine Building a Computer Model of Learning Classical Mechanics, Jude W. Shavlik and Gerald F. DeJong, University of Illinois at Urbana-Champaign Persuasive Argumentation in Resolution of Collective Bargaining Impasses, Katia Sycara-Cyranski, Georgia Institute of Technology The Interaction of Lexical Expectation and Pragmatics in Parsing Filler-Gap Constructions, Michael K. Tanenhaus and Laurie A. Stowe, University of Rochester; Greg Carlson, Wayne State University Predicting Conversational Reports of a Personal Event, Yvette J. Tenney, Bolt Beranek and Newman, Inc. Thematic Knowledge, Episodic Memory and Analogy in MINSTREL, A Story Invention System, Scott R. Turner and Michael G. Dyer, University of California, Los Angeles Spatial Inferences and Discourse Comprehension, Karl F. Wender and Monika Wagener, Technische Universitat Braunschweig Cognitive Processing Strategies for Complex Addition, Keith F. Widaman, David C. Geary, Pierre Cormier, Todd D. Little, University of California, Riverside ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Mon Aug 26 15:45:35 1985 Date: Mon, 26 Aug 85 15:45:15 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a004135; 13 Aug 85 3:26 EDT Date: Mon 12 Aug 1985 23:02-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #109 To: AIList@SRI-AI Received: from rand-relay by vpi; Tue, 13 Aug 85 21:26 EST AIList Digest Tuesday, 13 Aug 1985 Volume 3 : Issue 109 Today's Topics: Queries - Phone Numbers & AI Workstations, Review - AI Report Aug 1985, Literature - Recent Articles ---------------------------------------------------------------------- Date: Thu, 8 Aug 85 10:53:02 cdt From: Raj Doshi Subject: need phone numbers... Could some one please give me the PHONE-NUMBERS and E_MAIL_ADDRESSes of: ============= ================ 1. Dr. Elliot Solloway, YALE 2. Dr. Michalski (machine Learning book) 3. Dr. Lewis Johnson, (author of PROUST, ICAI-system Ph.D. Thesis at YALE.) Thanks very very much in advance. - Raj Doshi Grad Student, University Of Minnesota. ------------------------------ Date: Thu, 8 Aug 1985 18:04 EDT From: JIN%MIT-OZ@MIT-MC.ARPA Subject: AI Workstations [Forwarded from the MIT bboard by SASW@MIT-MC.] A friend of mine at Stanford CAD group is trying to choose a lisp machine for his lab. The candidates are: 3600, TI, Dandelion, and Tektronix. If you know any pros and cons or any literature comparing these machines, let me know. I would appreciate it. ------------------------------ Date: 9 Aug 1985 01:43-EST From: leff%smu.csnet@csnet-relay.arpa Subject: AI Report Aug 1985 Summary of Artificial Intelligence Report, Volume 2, No. 8 August 1985 o - Evaluation of Five Commercial Expert Systems Tools related to "Evaluating the Existing Tools for Developing Knowledge-Based Systems: by Mark H. Richer, Knowledge Systems Laboratory, Stanford University, Report NO. KSL 85-19 o - General Motors and the Unversity of Michigan have formed a Center for Machine Intelligence headed by Lynn Conway o - Proctor and Gamble and NYNEX have invested 3 million and 4 million in Teknowledge respectively o - Ted Harnett is now President and CEO of Quintus. He was previously at Micro Focus o - announcement of MacKit for writing expert systems on Macintoshes o - Kurzweill has announced production and marketing of a speech recognizer with a vocabulary of 1000 words o - Sperry - TI deal o - The Geneval Laboratories of Battelle Memorial Institute have completed a 2000 page survey of computer vision hardware o - Announcement of index to Scientific Data Link's AI memos o - bibliography by Harry Llull Review of: %A Paul Harmon %A David King %T Expert Systems: Artificial Intelligence in Business %I John Wiley & Sons ------------------------------ Date: 10 Aug 1985 02:25-EST From: leff%smu.csnet@csnet-relay.arpa Subject: Recent Articles %A Christine McGeever %T Symantec Ships Its First Product %J InfoWorld %D JUL 22, 1985 %V 7 %N 29 %P 20 %K Gary Hendrix Symantec %X Symantec, a company promising a significant AI product and which has hired Gary Hendrix, has shipped its first product. That product is a Lotus 1-2-3 which has nothing to do with AI. The "blockbuster" AI product is still planned. %T Teknowledge Gets $4M Cash Infusion %J Electronic News %D JUL 22, 1985 %V 31 %N 1559 %P 31 %K Proctor and Gamble %X Proctor and Gamble has invested 4 million dollars for Teknowledge in exchange for slightly more than 10 per cent ownership. %A Barbara Kellam-Scott %T Harvard Law School Computerizes the Paper Chase %J Hardcopy %D JUL 1985 %P 19 %V 14 %N 7 %K DEC %X Harvard Law School is automating their Legal Services Clinics. They have plans to include an expert system to assist lawyers in handling these cases. Digital Equipment has contributed to this program. %A Clara Y. Cuadrado %A John L. Cuadrado %T Prolog Goes to Work %J MAG4 %P 151-159 %K Symbolics Al Despain Yale Patt Berkely %X Al Despain and Yale Patt of Berkeley have achieved 425,000 LIPS using a custom designed processor. Symbolics has achieved 100,000 LIPS using custom microcode. Discusses general issues of Prolog in the contex of a maze traversing system. Also discusses the Japanese Fifth Generation project. %T Book Reviews %J MAG4 %X reviews of J. R. Ennals, Beginning Micro-Prolog and K. L. Clark and F. G. McCabe's Micro-Prolog Programming %P 49-60 %A Robert Kowalski %T Logic Programming %J MAG4 %P 161-177 %K Prolog %X general article on prolog, similar to other articles on the same subject by the same author appearing in other publications. %T Programming Environment with AI Module %J MAG1 %P 386 %X Commodore 64 Superforth Parsec Research %K Describes Superforth 64 + AI, a system to allow you to write expert systems on the Commordore 64. It supports antecedent and consequent reasoning ------------------------------ Date: 10 Aug 1985 02:20-EST From: leff%smu.csnet@csnet-relay.arpa Subject: bibliography These definitions are used for this AI bibliography and for the next submission which covers computer vision/ robotics articles. [I have passed the vision bibliography on to Vision-List@AIDS-UNIX. AIList readers can get a copy from AIList-Request@SRI-AI.ARPA, or by FTPing file vision.bib on SRI-AI. -- KIL] D MAG2 International Journal of Robotics Research\ %V 3\ %N 4\ %D Winter 1984 D MAG3 Journal of Mechanism, Transmission and Automation in Design\ %V 107\ %N 1 Special Issue\ %D MAR 1985 D BOOK14 Intelligent Robots and Computer Vision: Proceedings of the Society of Photo-Optical Instrumentation Engineers\ %D NOV 5-8 1984\ %E D. P. Casasent\ %E E. L. Hall D MAG3X Journal of Parallel and Distributed Computing\ %V 2\ %N 1\ %D FEB 1985 D MAG4 Byte\ %V 10\ %N 8\ %D AUG 1985 D MAG5 Industrial and Process Control Magazine\ %V 58\ %N 3\ %D MAR 1985 D MAG6 Aerospace America\ %V 23\ %N 4\ %D APR 1985 D MAG7 Computer Design\ %V 24\ %N 3\ %D MAR 1985 D MAG8 Manufacturing Engineering\ %V 94\ %N 4\ %D APR 1985 bib bibliography entries: ____________________________________________________________________________ %A J. A. Reggia %A B. T. Pericone %A D. S. Nau %A Y. Peng %T Answer Justification in Diagnostic Systems 1. Abductive Inference and its Justification %J IEEE Transactions on Biomedical Engineering %V 32 %N 4 %D 85 %P 263 %A J. A. Reggia %A B. T. Pericone %A D. S. Nau %A Y. Peng %T Answer Justification in Diagnostic Systems 2. Supporting Plausible Justifications %J IEEE Transactions on Biomedical Engineering %V 32 %N 4 %D 85 %P 268 %A K. Marik %T Customers Requirements of Natural Language Systems %J International Journal of Man-Machine Stuides %V 21 %N 5 %D NOV 84 %P 401 %A Stephen W. Oxman %T Expert Systems Represent Ultimate Goal of Strategic Decision Making %J Data Management %V 23 %N 4 %D APR 1985 %P 36 %A J. R. Ennals %T Beginning Micro-Prolog %I Harper & Row %C New York %D 1984 %X $15.95 %A K. L. Clark %T Micro-Prolog: Programming in Logic %I Prentice-Hall %C Englewood Cliffs, NJ %D 1984 %X $18.95 %A Tim Johnson %T Natural Language Computing: The Commercial Applications %I Ovum Ltd %C London %X Price $395.00 %A Ahrens, U. %A Schmidt, U. %T Advanced Programming Process for Industrial Robots %J Industrie-Anzeiger %V 106 %N 103/104 %D Dec. 28, 1984 %P 42-43 %A F. Bouille %T The 'HBDS' Database Model Kernel of a Structured Data Base System. Making Databases Work %J IEEE Proceedings of Trends and Applications %D 1984 %P 324-331 %A M. B. Cooper %A A. L. Kidd %T A Man-Machine Interface for an Expert System %J British Telecom Technology Journal %V 3 %N 1 %D JAN 1984 %P 112-115 %T Development Tool Debuts %J Computer World %V 19 %N 11 %D MAR 18, 1985 %P 49 %A R. Goering %T Do-It-Yourself Development Tools Speed AI Applications %J Computer Design %V 23 %N 14 %D DEC 1984 %P 29-32+ %A T. Huggins %T AI Systems Made Simple %J Informatics %V 6 %N 1 %D JAN 1984 %P 11-13 %A P. Hunter %T User Shells Sets Experts Wrangling %J ComputerWeekly %N 949 %D FEB 7, 1985 %P 28 %A V. P. Kobler %T Overview of Tools For Knowledge Base Construction %J International Conference on Data Engineering %I IEEE %C Los Angeles, Ca %D 1984 %P 282-285 %A J. R. Lineback %T Lisp Machine Provides a Shell for Industrial AI Applicaitgons In One of First Expert Systems To Go To Work %J Electronics Week %V 57 %N 2 %D 1984 %P 33 %A C. Maioli et al. %T Prototypes of Expert Systems for a Friendly Man Machine Interaction. User Termianls for Information/Communication Systems %J 31st International Congress on Electronics. Proceedings %D 1984 %P 35-42 %A A. Mehta %T ICL Claims Its Shell Cracks Expert Problem %J ComputerWeekly %N 943 %D DEC 13, 1984 %P 10 %A M. Merry %T Apex-3: An Expert System Shell for Fault-Diagnosis %J GEC Journal of Research %V 1 %N 1 %D 1983 %P 39-47 %A W. Rauch-Hindin %T AI Tool on PC bridges Expert to Novice Gap %J Systems and Software %V 3 %N 7 %D JUL 1984 %P 42+ %A S. E. Savory %T An Introduction to Artificial Intelligence with a Description of the Tool System, 'Twice' %J COMPAS 1984 Software as a Prduct %P 1057-1064 %A M. Schindler %T PC Software Tool Designs Small Expert Systems for Any Field %J Electronic Design %V 32 %N 12 %D JUN 14, 1984 %P 42-44 %A M. Schindler %T Expert Systems %J Electronic Design %V 33 %N 1 %D JAN 10, 1985 %P 112-114 %A M. STephens %A K. Whitehead %T The 'Analyst,' An Expert Systems Approach to Requirements Analyst %J European Seminar on Industrial Software Engineering and the European Workshop on Industrial Computer Systems. Proceedings % D 1984 %P 189-206 %T TI announces Expert System Software Tool %J Computerworld %V 18 %N 32 %D AUG 6, 1984 %P 59+ %A R. S. Wall %T Industrial Strength Knowledge Representation %J Third Annual International Phoenix Conference on Computers and Communications Proceedings %I IEEE %C Phoenix %D 1984 %P 6-10 %A R. C. Waters %T The Programmer's Apprentice: Knowledge-Based Program Editor %J Mini/Micro Northeast Computer Conference and exhibition %D 1984 %A Anthony J. Frisai %T AI Market Prospects are Good %J MAG5 %P 41 %A Dr. Richard A. Herrod %A Barbara Papas %T Artificial Intelligence Moves Into Industrial and Process Control %J MAG5 %P 45 %T I&CS Guide to AI Products %J MAG5 %P 50 %A Thomas E. Murphy %T Setting Up an Expert System %J MAG5 %P 54 %A A. C. Buffalano %T Expert Systems for the Military %J MAG6 %P 40 %A P. G. Freck %A R. P. Bonasso %T Drawing a Clear Picture of the Battlefield %J MAG6 %P 46 %A S. A. Vere %T Deviser- An AI Planner for Spacecraft Operations %J MAG6 %P 50 %A D. I. Smith %T CATS Precursor to Aerospace Expert Systems %J MAG6 %P 54 %A L. V. Filosofov %T Dynamic Recognition with Internal Instruction %J Engineering Cybernetics %V 22 %N 3 %D MAY-JUN 1984 %P 134 %A N. Mokhoff %T AI Techniques Aim to Ease VLSI Design %J MAG7 %P 33 %K AIENG %A H. J. HIndin %A S. F. Shapiro %T Speech Recognition Produces Natural Interface %J MAG7 %P 59 %A Z. L. Rabinovich %T Machine Intelligence and Fifth Generation Computer Structures %J Cybernetics %V 20 %N 3 %D MAY-JUN 1985 %P 426 %A W. Dilger %A W. Womannn %T The METANET: A Means for the Specification of Semantic Networks as Abstract Datatypes %J International Journal of Man-Machine Studies %V 21 %N 6 %D DEC 1984 %P 463 %A Michael B. First %A Lynn J. Soffer %A Randolph A. Miher %T QUICK (quick Index to Caduceus Knowledge) Using the Internish/Cadaceus Knowledge Base as an Electronic Textbook of Medicine %J Computers and Biomedical Research %V 18 %N 2 %D APR 1985 %P 137 %A Bernard Huet %T Semantic Modelling of Biological Sub-Systems by a Multilevel Control Structure Concept %J Kybernetes %V 14 %N 2 %D 1985 %P 93 %A J. A. Chester %T Artificial Intelligence-Is MIS Ready for the Explosion %J Infosystems %V 32 %N 4 %D APR 1985 %P 74 %A R. K. MIller %T Artificial Intelligence - A New Tool for Manufacturing %J MAG8 %P 56 %K AIENG %A R. P. Bergstrom %T AI - Fad with a Future %J MAG8 %P 65 %A Susan Walton %T Fighting Fire with Computers %T Technology Review %V 88 %N 6 %P 68-69 %D AUG/SEP 1985 %K forestry forest fires %X describes an expert system in a portable computer which is used to help fire fighters in determining how to fight forest fires. This system has been used to settle a law suit in Australia, to determine the appropriate amount of staffing for fire systems and has successfully predicted the area where a fire will occur and its path and duration of travel. ------------------------------ End of AIList Digest ******************** From csvpi@vpics1 Mon Aug 26 14:37:22 1985 Date: Mon, 26 Aug 85 14:37:09 edt From: csvpi@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a001457; 17 Aug 85 7:17 EDT Date: Wed 14 Aug 1985 21:46-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #110 To: AIList@SRI-AI Received: from rand-relay by vpi; Sun, 18 Aug 85 03:41 EST AIList Digest Thursday, 15 Aug 1985 Volume 3 : Issue 110 Today's Topics: Queries- Master Bibliography & OPS5 for Symbolics, Bindings - Friedman & Johnson, AI Tools - IBM Prolog and Expert Systems Development, Logic Programming - Opinion, Expert Systems - Definition & Database Systems & Rule Induction ---------------------------------------------------------------------- Date: Tue 13 Aug 85 09:28:58-PDT From: Mike Dante Subject: Master Bibliography I suspect that I am not the only one reading this BB who finds some of the submissions less than completely understandable due to lacunae in my own background. Hence a suggestion: Would it make sense to establish and maintain a bibliography (hopefully annotated), whose existence and address would be mentioned in the header of the AIList Digest? Then when someone like myself wanted to understand more, he or she could FTP a copy of the bibliography and with a little study, at least understand the terminology. ------------------------------ Date: Wed, 14 Aug 85 07:59:12 edt From: Martin Lee Schoffstall Subject: ops5 for symbolics does anyone have any pointers to an OPS5 for the symbolics3600? thanks, marty schoffstall schoff%rpics.csnet@csnet-relay ARPA schoff@rpics CSNET seismo!rpics!schoff UUCP martin_schoffstall@TROY.NY.USA.NA.EARTH.SOL UNIVERSENET RPI Computer Science Department Troy, NY 12180 (518) 271-2654 ------------------------------ Date: Tue, 13 Aug 85 15:28:07 cdt From: Raj Doshi Subject: want address of JPL... Hi, would anyone know the address of a Dr. LEN FRIEDMAN who used to (or still is) with (JPL) Jet Propulsion Laboratory ?? I have the old address. I sent a letter, but it was returned to me by U.S.Mail authorities. I know for sure that this was his address in June 1983. The old address was : Dr. Len Friedman =========== Automated Problem Solving Jet Propulsion Laboratory California Institute Of Technology 4800 Oak Grove Drive SUITE # 278 Pasedena, CA 91109 Would anyone know his PHONE-Number ????? Would anyone know his new address (surface or email) ?? Thanks very very much in advance. Please respond directly to me. Thanks again. --- raj doshi, University of Minnesota doshi%umn.csnet@csnet-relay.arpa ------------------------------ Date: 13 Aug 1985 0855-PDT (Tuesday) From: Johnson@ISI-VAXA Subject: binding and request I am told that a request for my current address appeared on ailist recently. My arpanet address is johnson@isi-vaxa; my mailing address is W. Lewis Johnson USC / Information Sciences Institute 4676 Admiralty Way Marina del Rey, CA 90292-6695 Also, could you please add my name to the recipients of ailist? Thanks. Lewis ------------------------------ Date: Sun, 11 Aug 85 10:22 P From: Henry Nussbacher Subject: IBM announces Prolog and Expert Systems Development VM Programming in Logic 5785-ABH One Time Charge Lang: Assem, REXX "VM Programming in Logic is an IBM implementation of the PROLOG programming language. It is suited for the research and development of applications in artificial intelligence including: expert systems, automated deduction,..." "VM Programming in Logic provides the following features: - Debugging Facilities - Communication with VM/SP ..., SQL/DS ..., and LISP/VM... (Note: the use of SQL/DS and LISP/VM are optional.)" The Availability is Sept. 6, 1985. Documentation: PDOM SH20-6541 Expert System Consultation Environment/VM 5798-RWP OTC or Monthly Expert System Development Environment/VM 5798-RWQ OTC or Monthly Language: PASCAL/VS "These two complementary program offerings provide the facilities for developing and executing expert systems. Expert System Development Environment/VM is used to 'build' knowledge bases. Expert System Consultation Env./VM is used to 'consult' those knowledge bases. ..." "These program offerings provide the following features: - English-like rules - Specialized editors with automatic checking to facilitate the entry and modification of knowledge base objects. - Explanation during consultation: 'Why?' provides a logical explanation for a certain request; 'What?' provides a more detailed explanation of the question being asked. - Debugging support. - Two inference processes: forward chaining, backward chainging. - Online help." "The Expert System Development Environment/VM requires the Expert System Consultation Environment/VM. Once the knowledge base (set of rules) has been developed, it can be replicated and used by the Expert System Consultation Environment/VM without the presence of the Expert System Development Env./VM." Availability: Sept. 6, 1985 Documentation: Gen. Info. Man. GH20-9597, ESDE Install: SH20-9607, ESDE User Guide SH20-9608, ESDE Ref. Man. SH20-9609. ESCE Install: SH20-9605, ESCE User Guide SH20-9606. These are supported out of Irving Texas. From Announcement Letter 285-284, August 6, 1985. ------------------------------ Date: Wed, 14 Aug 85 00:40:17 EDT From: Carl E. Hewitt Subject: Prolog will fail as the foundation for AI; so will LOGIC as a PROGRAMMING Language Prolog (like APL before it) will fail as the foundation for Artificial Intelligence because of competition with Lisp. There are commercially viable Prolog implementations written in Lisp but not conversely. LOGIC as a PROGRAMMING Language will fail as the foundation for AI because: 1. Logical inference cannot be used to infer the decisions that need to be taken in open systems because the decisions are not determined by system inputs. 2. Logic does not cope well with the contradictory knowledge bases inherent in open systems. It leaves out counterarguments and debate. 3. Taking action does not fit within the logic paradigm. [Carl also sent this message to the philosophy of science mailing list (Phil-Sci-Request%MIT-OZ@MIT-MC), and it has triggered several responses about Prolog/Logic Programming/AI. I am happy to let Phil-Sci carry the discussion, although it could just as easily have fit within AIList or the Prolog digest. For a good elaboration of Carl's thesis on open systems (networks, banking systems, nondeterministic distributed systems, etc.), see his article in the April '85 issue of BYTE. It's interesting reading, as are most of the articles in this special issue on AI. (The April Fool's What's Not column on pp. 96-97 is fun too.) -- KIL] ------------------------------ Date: Mon, 12 Aug 85 10:07 P From: Henry Nussbacher Subject: Expert System definition vs Database Systems I have been reading over the definitions of what an expert system is and isn't and I have seen in many of the comments that an expert system needs to be able to learn as it continues. Somehow, I have always felt Expert Systems to be glorified database systems. A database system gains more information as you add data to it. Th common example of Expert Systems (in my opinion) is the DOCTOR program: 1) Does the patient have a fever? Y 2) Has the patient vomitted in the past 24 hours? Y 3) Are the pupils dilated? N 4) etc... The AI program asks questions and based on the answers, determines what future questions to ask. In the end it narrows it down and comes up with a diagnoses based on the results of the questions. But I know of many database packages where a question in the form of: FIND FEVER > 100 & VOMIT = YES & DILATED = NO DISPLAY ALL My question is: What distinguishes the database search and display interface from an AI Expert System? Hank ------------------------------ Date: Fri 9 Aug 85 18:27:09-EDT From: SRIDHARAN@BBNG.ARPA Subject: Rule Induction and Expert Systems Masinter narrows it too far, making it anthropocentric. During the Machine Learning workshop at Allerton, IL, I heard Don Michie talk about the efforts of one of his friends. This effort involved building an ES to analyze EKG charts. They built two systems, one pretty closely following the Expert System methodology that Larry talked about. The other system was constructed by using a rule-induction technique, giving it a set of input charts and their analyses. The induction technique was biased toward a set of useful features for rule formation. The comparison of the two systems, yielded the conclusion that the induced-rule set outperformed the other in terms of both speed of execution and quality of results. One might admit so called expertise may be included in such weak forms as the bias given to the induction technique. Nature provides constraints for scientific theories. Some of us would like to tap into this ultimate source of "expertise" rather than stick to the derived expertise of humans. Masinter's description could be broadened to include this, if each occurrence of "expert" is not necessarily viewed as a human expert. P.S. Note that with the induced rule-set the system might be capable of explaining the rules themselves, by refering back to the cases. Another case serves as reference example. Larry and I both particpated in the Dendral and Meta-Dendral efforts at Stanford. The latter effort was aimed at rule formation for a class organic molecules; rules to characterize how bonds break in a mass spectrometer. Validated rules for several class of molecules were formed and incorporated into Dendral. Here both nature and human experts participated, but the human experts did not construct the rules. ------------------------------ End of AIList Digest ******************** From csvpi@vpics1 Mon Aug 26 14:42:20 1985 Date: Mon, 26 Aug 85 14:42:02 edt From: csvpi@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a000604; 17 Aug 85 3:57 EDT Date: Fri 16 Aug 1985 10:29-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #111 To: AIList@SRI-AI Received: from rand-relay by vpi; Sat, 17 Aug 85 21:22 EST AIList Digest Friday, 16 Aug 1985 Volume 3 : Issue 111 Today's Topics: Symbolic Computing - Definition ---------------------------------------------------------------------- Date: Sun, 11 Aug 1985 21:55 PDT From: DAVIES@SUMEX-AIM.ARPA Subject: What is symbolic computing? PARSYM is the new netwide mailing list for parallel symbolic computing. One of the first messages to PARSYM asked just what is symbolic computing anyway. I thought the readers of AIList might be interested to see what PARSYM came up with. -- Byron Davies (PARSYM-Request@SUMEX-AIM.ARPA) ------------------------------ Date: Mon, 8 Jul 85 09:15:00 pdt From: coraki!pratt@Navajo (Vaughan Pratt) Subject: What is symbolic computing? I suppose with quarter of a century of Lisp experience we should understand by now what symbolic computing is. Indeed everyone appears to be quite happy to talk about symbolic computing as though the concept had the unanimous blessing of the last three centuries of mathematics. Yet I have been unable to find in the literature a definition of the concept that matches the facts. Discussion based on the premise that the topic is well-defined when it isn't is apt to run at cross-purposes. I therefore challenge this list to agree on a definition of symbolic computing. Even if this challenge cannot be met (as I expect will be the case), at least people will have been made aware of the variety of definitions and will know to make allowance for this variety. -v ------------------------------ Date: Mon, 15 Jul 85 15:34:33 EST From: munnari!dmtadel.oz!crw@seismo.CSS.GOV (charles watson) Subject: Applicative Symbolic Programming Firstly I consider symbolic computing, to be abstract and non-numeric, dealing with structured objects that match human concepts and images. Secondly, I'd like to argue the case for side-effect free programming. I feel that the side-effects of much real-world software are the result of poor programming practice. John McCarthy's (1959) LISP is sufficient. All the features added in the name of efficiency have encumbered their compilers and interpreters. Except for debugging, side effects have been tolerable in the past. For highly concurrent systems of the future, side-effects would be catastrophic. I accept that the cost of re-writing software is a strong reason for downward compatability to existing non-applicative code, but redesign in a sound paradigm would take less effort than conversion. The machine architecture I'm working on will allow setq to avoid re-evaluating S-exprs but not for explicit register assignment; it will allow replac* to append non-circular lists, but not to point back to any object in the ancestral tree. There are those who still believe that the shortest distance between two pieces of software is a GOTO statement. In my experience real-world software is easier to develop and maintain in the structured paradigm. The last project I was involved in used an applicative hierarchical specification of IC designs to drive a silicon compiler. The problem would have otherwise been intractable. Also, processors can be custom designed for symbolic processing and replicated for the cost of about $10 each (ignoring the development cost). A system with thousands of these processors, could cost the same as a Symbolics 3600. Arguments such as "this side-effect riddled feature gives a 20% performance improvement" would be irrelevent to the cost-effectiveness of such a system. ------------------------------ Date: Tue, 16 Jul 85 07:46:47 pdt From: coraki!pratt@Navajo (Vaughan Pratt) Subject: Definitional Issues From: munnari!dmtadel.oz!crw@seismo.CSS.GOV (charles watson) Firstly I consider symbolic computing, to be abstract and non-numeric, dealing with structured objects that match human concepts and images. Excellent! A first attempt (on PARSYM) to define symbolic computing. Let's try it out. 1. What does "abstract" mean? One definition I like a lot is "authorized." Computation is abstract when it depends only on what is authorized by the documentation. Is that what you meant? If so, why is that in the definition of "symbolic" computing, as opposed to other kinds of computing? If not, then what do you mean? 2. Lisp has numbers. Does this rule out Lisp as a symbolic language? 3. In PROLOG without equality an inefficient but convenient way to represent natural numbers is symbolically: 3 = (S (S (S 0))) and so on. How do you reconcile "symbolic" and "nonnumeric" for such a case? 4. What is an example of a structured object *not* matching human concepts and images? I find it hard to conceive of or imagine such a thing. This requirement is surely more appropriate for a definition of AI computing than symbolic computing. In short, I see neither neither the rationale for nor the application of any part of this definition. Secondly, I'd like to argue the case for side-effect free programming. ... For highly concurrent systems of the future, side-effects would be catastrophic. Speaking as one highly concurrent system trying to side effect another, I hope I haven't thereby caused a catastrophe. And on behalf of human society, another highly concurrent system, it would certainly be interesting, and surely catastrophic in some sense, if we ceased to have side effects on each other. Whether you have, or for that matter can identify, side effects is very dependent on your particular computational paradigm, e.g. the distinction between functional and imperative programming. As soon as one starts to explore other paradigms appropriate for concurrency, e.g. dataflow (of particular interest to me), the concept of side-effect free programming becomes either irrelevant or meaningless. About the only sense one could make of it in dataflow would be if one introduced ESP for processes, i.e. communication by unseen channels. Sorry not to be more constructive here. For more constructive remarks in the above spirit see my POPL-83 paper "Five Paradigm Shifts in Programming Language Design and their Application to Viron, a Dataflow Programming Language" After having taken off a couple of years helping out with getting workstations out to people I am just now returning to academia to continue the design and implementation of Viron, or something resembling it (in addition to continuing my work on fonts, a side-effect (hak coff) of my working at Sun). If people would be interested I'd be happy to make occasional short contributions to this column expressing the general philosophy behind Viron, which is very much a parallel and abstract programming language. Since Viron processes don't have a notion of internal state (how do you define "the state" of a process consisting of an ocean of ships each loaded with microprocessors with cycle times measured in nanoseconds, where "simultaneous" is both physically and practically undefined?) one has to define "side effect" in a way that does not depend on the notion of state - in this sense Viron is free of any state-based notion of side effect. Whether Viron could be called "symbolic" depends on whether we ever find a workable definition of "symbolic," but it should pass almost any plausible definition that does not rule out numeric computation and that does not specify implementation or representation details. -v ------------------------------ From: eugene@AMES-NAS.ARPA (Eugene Miya) Date: 16 Jul 1985 1736-PDT (Tuesday) Subject: Re: PARSYM Digest V1 #1: RE: What is symbolic computing? > From: coraki!pratt@Navajo (Vaughan Pratt) > Indeed everyone appears > to be quite happy to talk about symbolic computing as though the > concept had the unanimous blessing of the last three centuries of > mathematics. I work with people who crunch numbers, permit me to play devils advocate (I do support research on symbolic processing): the only people who have given their blessing is the LISP community. Let's not close our eyes to that fact. > I therefore challenge this list to agree on a definition of symbolic > computing. I realize this is rehashing hallway arguments we have all had: please define that which is "not" symbolic computing and why we should make a distinction in these two types of parallel computing. After all isn't crunching a number the same as manipulating a symbol, and aren't we possibly creating artificial distinctions of computing types (symbolic and numeric)? I like the idea of academic discussions of this nature. Work can get too serious at times. I also would like to point out that I just returned from SU and I have seen copies of Dr. Pratt's TRs on new thoughts for concurrency models. --eugene miya NASA ------------------------------ Date: Fri 19 Jul 85 21:25:35-CDT From: Mayank Prakash Subject: Re: What is symbolic computing? Here's another attempt at it. First of all, I think that the terms symbolic computing and numerical computing are different modes of computing rather than mutually exclusive taxonomical categories. Then, numerical computing is the mode when the major data elements are numerical and one is interested in changing the numerical values of these data elements. That is, both the input to and the output from the program are mainly numerical. In symbolic computing, one is interested mainly in manipulating structures. That is, both the input and the output to the program are structures. Note that in this definiton one mode of computing does not exclude the other. In fact, most programs do some of each. It is the predominant activity of the program that determines it's mode. One could look at it from a lower level as well. The memory cells in the computer's data memory (as opposed to the instruction memory) contain binary values. If they are mostly interpreted as representing numbers, and the majority of operations that are carried out on them are numerical, i.e., add, subtract, multiply, shift etc. their values, then the program is a numerical mode program. If they are mostly pointers to other cells in memory, and the operations on them are mainly follow the pointers along, modify them to point to some other cells etc., then the program is a symbolic mode program. A characteristic that generally distinguishes the languages for the two kinds of programs is memory allocation. The languages developed for numerical processing have mostly static memory allocation schemes. By that I mean that the data memory is allocated to a procedure (or a function, or block, whatever you want to call it) upon entry, and released upon exit. Generally, though not always, the procedure does not (and in most cases, can not) change its data memory. In contrast, symbolic processing languages have dynamic memory allocation with attendant garbage collection. This is necessary for symbolic computing since in this case one is dealing with structures, which are essentially pointers pointing at each other in various ways, and since the main activity here is manipulating these structures, i.e., releasing and allocating pointers. Admittedly these are somewhat vague definitions, but I hope that this posting will at least spur a discussion on the subject. - mayank. ------------------------------ Date: Thu, 25 Jul 85 22:43:43 edt From: Tom Blenko Subject: What is symbolic computation? Vaughn's question is an interesting one. My proposal is that numerical computation is performed over a flat domain, while symbolic computing permits computation over terms which are partially bound or instantiated. Binding is used in a broad sense here, and subsumes type declaration and generic instantiation, as well as the conventional notion of variable binding. According to this scheme, FORTRAN is almost purely numerical because it allows variables to be bound only to constants (although a form of compile-time co-referential binding is possible using EQUIVALENCE). FORTRAN is not purely numerical because variables are typed. Type declaration is a form of binding under the notion of binding described above, although an especially weak one because all type declarations must be made at compile time. The class of nearly-numerical languages (those with flat domains plus some support for typed variables) can be expanded somewhat by including languages with slightly more powerful type mechanisms, i.e., those which support discriminated unions or procedure name overloading. For the purposes of discussion, I'd be willing to refer to all of these as languages which support numerical computation exclusively. This seems like a reasonable approximation because their competence as symbolic languages is both weak and well known. Languages which permit various forms of abstract data type or generic procedure definition would be classified as partially-symbolic because they each support some form of run-time partial binding of variables. Representative examples in this class are CLU, ADA, and SMALLTALK-80. This is the class currently of greatest interest to the imperative language people, and certainly there needs to be more work on what abstraction and type mechanisms are useful and can be implemented efficiently. My two choices to represent (nearly) fully-symbolic languages are PROLOG and LISP. In PROLOG, the binding mechanism, unification, can also be viewed as a type restriction mechanism, so that a variable becomes bound to a grounded term by successive application of type restrictions to (variable) subterms of intermediate bindings as the computation proceeds (reference for related work is Hassan Ait-Kaci's thesis, A New Model of Computation based on a Calculus of Type Subsumption). This identification of variable typing with more traditional notions of variable binding is precisely what I propose permits one to view symbolic computation in a coherent way. LISP is well-known, and it would be quite a task to persuade some that it is anything except the ultimate symbolic language. Clearly its binding mechanism allows the kind of partial binding which occurs naturally in PROLOG (although, of course, this could be said to be only one consequence of its excessive permissiveness). Let me mention two ways in which it differs from PROLOG, however, and might be viewed as a more powerful symbolic language. First, it allows variables to be bound as pointers to other variables. This is undeniably a powerful mechanism, although it makes for more complicated language semantics. It is also not a particularly good substitute for what is (arguably) the corresponding mechanism in PROLOG, specifically the co-referential binding resulting from the unification of variables. I say the two mechanisms correspond because the only way to do equivalencing in LISP is for variables A and B to both point to C, and for C to store the equivalenced value of A and B, which may be accessed by a dereference followed by an evaluation -- which leads to the second point. Many LISPs implement what might be termed a first-class recursive call to the interpreter. INTERLISP's evala, for example, allows any procedure to call the interpreter recursively on a LISP data structure with the binding environment of the computation completely specified in an argument to evala. Intuitively, this is a powerful mechanism for symbolic computation, and is moreover necessary for the rather awkward implementation of LISP equivalencing described above, since dereferencing is indistinguishable from evaluation (although Brian Smith has succeeded in separating the two in his definition of 3-LISP). The (second-class) recursive call implemented in most PROLOGs is more restricted because the environment of the calling procedure is unconditionally inherited by the called procedure (although proposals for more general approaches have been made). Many partially- and non-symbolic languages do not provide any recursive call. These are admittedly incomplete thoughts, and I'd be interested in any responses. I have not specifically addressed binding of parameters across procedure calls -- call-by-value and call-by-reference can be understood rather easily once the notion of variable creation is included, I think. I haven't though about the symbolic power of exotica such as thunks, and I suspect that macro expansion doesn't add much in the way of symbolic power. I'd be particularly interested in a coherent exposition of the relationships between what I've been proposing as the primary characteristic of symbolic computation (partial binding) and mechanisms such as pointer creation and dereferencing, and lazy or eager evaluation. For example, one might interpret PROLOG unification of variables as a lazy assignment of the source variable to the target variable, with evaluation of the source variable binding delayed indefinitely (this is correct because PROLOG variables are write-once). The obvious way to do dereferencing in PROLOG is through a "procedure call" or recursive call to the interpreter, except that the PROLOG interpreter treats bound and unbound variables differently, so that unbound variables "evaluate" to themselves both during unification and when used as parameters to recursive function calls (under what interpretation is this eager evaluation?). Another question is why one is normally tempted to categorize a language like C as non-symbolic, since it allows liberal pointer creation/dereferencing and thereby allows the same binding of variables to variables as LISP to be performed in a fairly direct fashion. (Note, however, that no recursive call to the interpreter is permitted). I'd be interested in any thoughts or comments the list might have. Tom BLENKO@ROCHESTER ---------------------------------------------------------------------- Date: Fri, 26 Jul 85 10:04 EDT From: Seth Steinberg Subject: Symbolic vs Numerical Computing We had better be careful here. Programs, not languages are either symbolic or numerical. If I write a numerical matrix inverter in LISP it is still a numerical program while if I write a MACSYMA-like symbolic algebra matrix inverter in Fortran it is a symbolic program. (Never mind why I would want to do the latter). Numerical programs make the vast majority of their decisions based on the values of terminal values which are stored in relatively homogeneous data structures. Symbolic programs make significantly more decisions based on examination of the structure of the data which may vary more freely. The exact implementation of runtime typed data may be part of the programming language (LISP or SMALLTALK types), assisted by the programming language (PASCAL records used with case or C struct unions), or it can be implemented in spite of the programming language (FORTRAN, possibly with a package like SLIP or ASSEMBLY language using the high bits for type and the rest for pointer). [Obviously, some languages make symbolic programming easier than others]. This definition is far from perfect so I'll propose a test. Try ranking a list of programs on the numerical<->symbolic scale. For example (probably not in the right order): Fast Fourier Transform Sparse Matrix Multiply Graph Coloring Algorithm Simple Expression Compiler ELIZA LISP Interpreter Simple Theorem Prover Algebraic Integrator Noun Phrase Parser Try out your own ordering. Where would you put in things like: FTP support for the Symbolics 3600 UNIX (kernel or cshell) MacPaint vs. MacDraw A Turing Machine I'd be interested to see how these would be ranked or whether it is meaningless to perform such rankings. Seth Steinberg SAS @ BBN-VAX ------------------------------ Date: Sat, 27 Jul 85 14:14:44 pdt From: coraki!pratt@Navajo (Vaughan Pratt) Subject: What is symbolic computing? From: Mayank Prakash Then, numerical computing is the mode when the major data elements are numerical and one is interested in changing the numerical values of these data elements. ... I do a lot of computing with: * complex numbers * polynomials over various fields, including the reals * vector spaces of various dimensions * linear transformations * survey maps, involving bearings, lot boundaries (expressed as lists of line segments), areas, etc. * outline fonts based on conic splines Now if these aren't examples of structured data I'm a monkey's uncle. Yet most of the time spent manipulating these structures goes into floating point operations. On the one hand this is certainly consistent with Mayank's observation that "most programs do some of each." On the other hand I don't see how to apply the test "predominant activity." Is this determined by the proportion of time spent on floating point operations? If so then does plugging in a floating point accelerator convert my program from a numerical to a symbolic one? Or is it determined by the number of calls to floating point routines in my programs? Most of my calls are to things like dot and matrix products. If [the memory cells] are mostly interpreted as representing numbers, and the majority of operations that are carried out on them are numerical, i.e., add, subtract, multiply, shift etc. their values, then the program is a numerical mode program. ... At this level the definition is vulnerable to the compiler. Given a vector of reals a powerful optimizing compiler may see fit to implement it either as a linked list (if the optimizer detects operations on the vector that amount to expanding or contracting the vector) or an array of contiguous locations (if there is much random access to the array). How does one classify a program that leaves such decisions to the compiler? ... symbolic processing languages have dynamic memory allocation with attendant garbage collection. When memory is allocated and released in LIFO order it is very efficient to allocate it off a stack. When the release order is unpredictable one resorts to a heap. How does this have anything to do with whether numeric data types are involved? -v ------------------------------ Date: Wed, 31 Jul 85 11:59 EDT From: Guy Steele Subject: Symbolic and numerical computing Symbolic programs: * laugh at themselves. * philosophize. * till the soil. * are featherless bipeds. Here's a more serious attempt: ALL computing applications are symbolic. All applications rely on processing data organized according to some structural discipline. This discipline may be trivial or exceedingly complex. There are typically certain invariants or axioms of the structure, and the operations on the structure, on which the processing relies: for example, that a tree is binary and balanced, and insertion and deletion maintain these invariants. A particularly large and important class of applications relies heavily on the axioms for rings and fields, particularly the ring of integers and the fields of real and complex numbers, and for such applications much of the computation is done with data structures and operations that are organized so as to obey these axioms (more or less, given the usual finiteness of the representations). Because these applications are so important, and the theory is well-understood and agreed-upon, special hardware accelerators for certain very complicated operations (such as multiplication) are the norm rather than the exception; but the presence or absense of such hardware has, to my mind, little bearing on the numericalness of the application. So I propose that an application be considered numerical to the extent that it relies on data structures obeying the axioms for rings or fields, however these data structures may be represented as bits. I would regard a LISP program operating on lists of NIL's as numerical if it were so organized as to treat these lists primarily as unary encodings of numbers, using routines to concatenate the lists (addition) and repeatedly self-concatenate (double, multiply), and so on. (Indeed, the SCHEME chips that I and others designed to directly interpret LISP code had no on-chip ALU to speak of, and the chips were tested on numerical applications using numbers encoded in exactly this way.) Is a program that relies on a group structure numerical? What if there is hardware to compute a*b quickly, where * is the group operator? Suppose the group were SU(16) or GF(16) instead of Z[2^16]? By the proposed criterion, any such program might be somewhat numerical, but less so than one using a ring or field. --Guy Steele ------------------------------ Date: Fri, 2 Aug 85 22:18:42 pdt From: coraki!pratt@Navajo (Vaughan Pratt) Subject: Re: PARSYM Digest V1 #8 From: Guy Steele ALL computing applications are symbolic. My position exactly. I particularly appreciated GLS's algebraic examples, which I thought were right on. Subject closed (as far as I'm concerned). -v ------------------------------ Date: Tuesday, 6 August 1985, 5:15 pm PDT From: Byron Davies Subject: What is symbolic computing? The cat is out of the bag. Symbolic computing is indeed all computing. PARSYM was designated the netwide mailing list for *symbolic* computing in order to broaden the domain of discourse rather than to restrict it to any particular branch of computing. I hope it won't be long before someone asks the analogous question about parallel computing -- and gets the *same* answer. -- Byron ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Mon Aug 26 14:20:45 1985 Date: Mon, 26 Aug 85 14:20:36 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a001800; 19 Aug 85 1:02 EDT Date: Sun 18 Aug 1985 21:06-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #112 To: AIList@SRI-AI Received: from rand-relay by vpi; Mon, 19 Aug 85 21:46 EST AIList Digest Monday, 19 Aug 1985 Volume 3 : Issue 112 Today's Topics: Query - GF11 Hardware, Information Science - Online Bibliographies, Expert Systems - Definition vs Database Systems, Logic Programming - Hewitt vs Prolog, Seminars - Design Expert Systems (CMU) & - Intelligent Design Completion (CMU), Call for Papers - Knowledge Represention (Proc. IEEE) ---------------------------------------------------------------------- Date: Sat, 17 Aug 1985 07:44 EDT From: "David D. Story" Subject: GF11 Hardware Has anyone out there looked at SIGARCH proceeding of last month. I was wondering what comments might be found on IBM's GF11. I heard of this machine years ago. I thought at that time it was for a group of problems General Foods had taken to the Yorktown Group. I see that they have added some new micro-code features resembling Hewitt's Actors and Semantic Massage for a society of experts. Though I dunno as of yet what is happening with AI at Yorktown it seems they've tried to get this under the hood. If the machine was meant for the task of computing chromodynamics seems to me that the word size would have been re-engineered since there is no mention of double or quad wording in the article. Whatta think ? Dave ------------------------------ Date: Sun, 18 Aug 85 12:46:07 pdt From: aurora!eugene@RIACS.ARPA (Eugene miya) Subject: Re: Master bibliography Excellent suggestion. The net's biggest problem is a lack of memory. ;-) I am trying something like this right now for parallel and distributed computing. Mine is ftp-able, and has over 5000 entries. It also has some copyright restrictions because I used several preexisting bibliographies [i.e., stand on the shoulders of giants...]. There is only one bibliography in the field larger than mine, but it's hardcopy, hard to use, but it has many more European papers. Mine's dynamic, useable with text formatters, and updateable. Keys and annotations, too. I suspect an AI bibliography will have two major problems: 1) AI is a much bigger field. 2) AI has more hype literature associated with it. If it were possible to moderate the technical content of the papers, you will succeed nicely. I have a separate bibliography for the "top ten" required readings in software engineering. I plan to update it yearly with a call for suggestions. Books will be booted in and out (I hope). Other minor problems: some work was received in Scribe bibliographic format: I decided on refer: ran on smaller machines, Unix more widely available, and so on. I had to write crude Scribe->refer translators. Getting people to help add, correct, and delete work is surpising difficult: everybody wants loaves of bread, but few want to do the work. The initial start is the hardest of course. Try to build off of others work if they will let you. --eugene miya NASA Ames Research Center emiya@ames-vmsb.ARPA ames!aurora!eugene on UUCP ------------------------------ Date: Sun, 18 Aug 85 00:35:00 edt From: BostonU SysMgr Subject: Re: Expert System definition vs Database Systems From: Henry Nussbacher ...Somehow, I have always felt Expert Systems to be glorified database systems.... 1) Does the patient have a fever? Y 2) Has the patient vomitted in the past 24 hours? Y 3) Are the pupils dilated? N 4) etc... But I know of many database packages where a question in the form of: FIND FEVER > 100 & VOMIT = YES & DILATED = NO DISPLAY ALL In the first place, differential diagnosis is both a good and a bad example. Bad because it is meant to provide a lot of structure that can be likened to a data-base query with boolean logic and good because it has been worked on a lot in AI and as you get into more details it starts to become more clear why the database approach isn't always powerful enough. Consider: In the first place, there are many, many diseases. A doctor doesn't attempt to know all of them. In fact, the questioning (in a doctor's mind) I believe starts with something more like: is this person in front of me about to drop dead? a lot of info has to be processed real fast and inaccurately (from a data base/strict machine point of view) to answer that and act on it. Ok, let's try it again: IF s/he has a fever AND s/he has been vomiting THEN (will s/he drop dead in a moment?) or FIND DISEASE WHERE FEVER & VOMIT & DEATH hmmm, doesn't work. Maybe that's all the patient is saying though. I guess we better find out if s/he's severely dehydrated, measure the fever, or maybe they just have a little food poisoning. Ok, try again: IF he has a fever AND he has been vomiting THEN he has malaria... wait a minute! there's no malaria around here...try again (darn, if s/he hadn't just fallen over I might have asked if s/he have been traveling in the tropics lately or eaten any jalisco cheese, now what do I do...) I think my point is, yes, it's kind of like a database query BUT WHO IS GENERATING THE QUESTIONS. I think your example weakens a lot once the first query is made, who decides what the second query is to be? The expert system of course. You are assuming some magic actor generating all these nice queries and inferences, get rid of that actor and try it again. -Barry Shein, Boston University ------------------------------ Date: Sat 17 Aug 85 10:38:41-PDT From: PEREIRA@SRI-AI.ARPA Subject: Hewitt's tirade against Prolog Carl Hewitt's message is based on several misconceptions: 1. (the least interesting one) All the so-called commercially viable Prolog systems in Lisp are not really Prolog systems written IN Lisp, but rather Prolog systems written FOR Lisp machines. Or is it that a microcode sublanguage or Lisp machine pointer-smashing operations are part of List as we know it? Without those machine-level operations, those Prolog systems would run too slow and use too much memory to be useful for serious Prolog programming. From the Prolog implementation point of view, what is important about the Lisp machines is not that they run Lisp, but that they can be microcoded and have good support for tagged data types and stack operations. 2. If the decisions (actions) of a system are not determined by its inputs, the system is nondeterministic. Nondeterminism in a system can be either an artifact of our incomplete knowledge (or lack of interest) of the detailed operation of the system; or it can be ``real physical'' nondeterminism. It would take us to far to discuss whether the second kind of nondeterminism is ``real'' or also an artifact. In any case, most uses of nondeterminism, say in models of concurrency, are of the first kind, and can be expressed appropriately in various temporal/dynamic logics. Admittedly, these are not Prolog, but then Common Lisp is not Lisp 1.5! (Prolog is 13 years old, Lisp 25). 3. The first logic course dictum ``from a contradiction one can conclude anything'' is getting in the way. Notice that the dictum says ``can'', not ``must''. There is an enormous difference between things that are in principle true and things that an agent knows to be true in a way that can affect its action. An agent might know ``p'' and ``not p'', but it might well never come to infer the dreaded totally unrelated ``q'' which IN PRINCIPLE follows from the contradiction. This inference might not happen either because of inference control mechanisms of the agent (eg. it uses the set-of-support strategy) or because the agent's logic is just TOO WEAK to conclude anything from a contradiction (vide Hector Levesque's work in the proceedings of the last AAAI). In any case, Horn clauses (the basis of Prolog) are too weak to represent contradictions... :-) 4. The question of whether ``taking action'' fits in a logic paradigm tends to be answered negatively after an hour's worth of consideration. If you persist for several years, though, this question becomes a source of insight on the relations between knowledge, state and action that is not available to those who started by dismissing the question after that initial hour. There is just too much work on logics of knowledge and action in AI and computer science for me to try to discuss it here. Some of this work has been applied to logic programming, either in the form of new logic programming languages based on temporal or dynamic logics or in representations of temporal reasoning and decision in, say, Prolog. 5. It is curious to see someone by implication defend Lisp as a language for expressing the taking of action! We know by now that the most difficult issue in ``reactive systems'' is not EXPRESSING action, but rather keeping it under control to prevent unwanted interactions. In this area, less is REALLY more, and highly complex languages like Lisp are just not suitable for the formal reasoning about programs that is needed to help us believe our programs do what we intend. To those who say ``...but we can keep to a carefully constrained subset of Lisp, use only message passing for interactions,...'' I will answer that the history of all large Lisp programs I know (and I think that is a representative sample) is littered with the brutalized corpses of constrained programming styles. Anyone who has looked at the current flavor mechanism in Zetalisp and its use in the window system will know what I mean... 6. Underlying Carl Hewitt's misconceptions is the old chestnut ``logic is static, systems are dynamic''. Any language, be it first-order logic or Lisp, is static; it is its USE which is dynamic (changes the state of communicating agents). A good analogy here is the use of differential equations to model dynamic systems in classical mechanics. The differential equations themselves do not say directly what happens when (they are ``static'' in Hewitt's jargon). It is their SOLUTIONS that tell us the sequence of events. Even the solutions are given as static objects (functions from an open interval of the reals to some space). Does anyone worry that such equations do not ``really'' describe the dynamic behavior of a system? Is it not possible to combine such ``static'' entities with an incremental solution procedure to build systems that actually control their (classical mechanical) environment? -- Fernando Pereira ------------------------------ Date: 15 Aug 85 15:05:59 EDT From: Mary.Lou.Maher@CMU-RI-CIVE Subject: Seminar - Design Expert Systems (CMU) A GENERATIVE EXPERT SYSTEM FOR THE DESIGN OF BUILDING LAYOUTS Ulrich Flemming Design Research Center & Department of Architecture Thursday, August 22 at 1:30 pm Adamson Wing, Baker Hall The talk will outline a generative expert system for the design of building layouts aimed at systematically enumerating layout alternatives while taking into account a broad range of criteria, a task to which the human cognitive apparatus is not particularly well suited. The system is roughly modelled after the DENDRAL system. In its most simple incarnation, it will consist of a generator able to generate all possible alternatives, a tester that evaluates these alternatives, and a control strategy that mediates between the two to help prune the search tree. What makes the generator so special is that it treats spatial relations between the objects to be allocated as the basic design variables in which the generation takes place. The completeness and non-redundancy of the generation have been established. The tester will be programmed to facilitate the addition and modification of the design knowledge incorporated in it. A tentative control strategy will be discussed. It is expected that for more complicated layout problems, the control strategy will have to be expanded into a genuine planner with at least two levels: 'hierarchical' and 'strategic', both of which will be outlined. ------------------------------ Date: 13 Aug 85 13:23:26 EDT From: Jeanne.Bennardo@CMU-RI-ISL1 Subject: Seminar - Intelligent Design Completion (CMU) Topic: Presentation of Wright Project Speaker: Can Baykan Place: DH3313 Date: Wednesday, August 14 Time: 10:00am - 11:00am An intelligent design completion system is a knowledge-based CAD system which provides a design environment and assists the designer in analyzing and synthesizing designs. For example, the designer may generate a partial design and have the system carry out a diagnostic evaluation, or complete the design. Such a system would be composed of two major components: a knowledge-base and a drafting system. The WRIGHT system is an interactive CAD system which the designer can use in representing, analyzing and generating kitchen designs. The goals in building such a system are to understand: 1. The architecture and components of a design-completion system. 2. The types of knowledge required for analyzing and synthesizing designs, Knowledge required for recognizing elements in a drawing generated by the designer, Knowledge required for recognizing design contexts, 3. The generation of form -a complex, structured object-, based on function -a diverse set of constraints from many sources. The application domain chosen for the WRIGHT system is kitchen design. ------------------------------ Date: 13 AUG 85 15:57-N From: ROSNER%CGEUGE51.BITNET@WISCVM.ARPA Subject: Call for Papers - Knowledge Represention, Proc. IEEE CALL FOR PAPERS Proceedings of the IEEE Special Issue on Knowledge Representation Guest Editors: M King, M Rosner, University of Geneva The special issue is scheduled for publication during the second half of 1986. You are invited to submit a 6-10 page extended abstract on any topic relevant to the current state of the art in Knowledge Representation. Deadlines: submission of abstracts: 30th September 1985 notification of acceptance: 30th December 1985 final copy: 15th February 1986 contact: ROSNER%cgeuge51@WISCVM.ARPA (bitnet) mcvax!cernvax!cui!rosner (usenet, eunet, uucp) M Rosner ISSCO, 54 route des Acacias, 1227 Geneva, Switzerland ------------------------------ End of AIList Digest ******************** From csvpi@vpics1 Tue Aug 27 01:00:51 1985 Date: Tue, 27 Aug 85 01:00:46 edt From: csvpi@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a001782; 26 Aug 85 0:29 EDT Date: Sun 25 Aug 1985 20:33-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #113 To: AIList@SRI-AI Received: from rand-relay by vpi; Tue, 27 Aug 85 00:39 EST AIList Digest Monday, 26 Aug 1985 Volume 3 : Issue 113 Today's Topics: Queries - Machine Learning Journals & Lisp to C & Hardware Choices For Running LISP & Modularity and Compositionality & Expert Systems in Psychiatry, Literature - Master Bibliography, Humor - Media Portrayal of Scientists, Games - Book Openings, Logic Programming - Hewitt's Reply to Pereira, Seminar - Partial Evaluation in Meta-Interpreters (SU) ---------------------------------------------------------------------- Date: Sun, 18 Aug 85 23:00:23 cdt From: Raj Doshi Subject: machine learning journals.. I (as a co-author) would like to publish a modest article which has something to do with MACHINE LEARNING. Can someone tell me the JOURNALs that are strictly for machine learning ???? Are there any such conferences. ???? Thanks in advance. --- raj doshi, University of Minnesota doshi%umn.csnet@csnet-relay.arpa [Gluck@SU-PSYCH announced a new Machine Learning journal in AIList V. 3, No. 106, Aug. 10. It will come out quarterly beginning January 1986. Executive Editor: Pat Langley; Editors: Jaime Carbonell, Ryszard Michalski, Tom Mitchell; Kluwer Academic Publishers, 190 Old Derby Street, Hingham, MA 02043, or call 617-749-5262. -- KIL ] ------------------------------ Date: 20 Aug 85 17:18:44 EDT (Tue) From: duke@mitre.ARPA Subject: Lisp to C We are looking for a Lisp to C translator, in the hope that it might help us move some Franz Lisp programs onto some parallel processor machines which currently lack Lisp. Does anyone know of such a translator? Duke Briscoe Mitre ------------------------------ Date: Fri, 23 Aug 85 12:09 EST From: Clarke Thacher Subject: Hardware Choices For Running LISP. Our computing center director has asked me to get some information and make recommendations on alternatives for supporting LISP at the University. Specifically, he has been getting researchers in several departments submitting requests for single user LISP workstations. He is reluctant to approve all of these requests, if a more economic solution is available. I would appreciate any pointers which people might be able to offer. We have an IBM 3081 with VM & CMS, and 3 Prime superminis. One of the Primes has the Salford LISP/Prolog system but I suspect that the researchers want more than that. Approximate costs would be appreciated. Clarke Thacher BITNET : UKC323@UKCC (606) 257-2900 72 McVey Hall Computing Center University Of Kentucky Lexington, Ky. 40506-0045 ------------------------------ Date: Wed 21 Aug 85 21:19:25-PDT From: Lee Altenberg Subject: Modularity and Compositionality A few issues back someone used the term "modularity" to refer to parts of a program. This leads me to ask, is there a precisely defined notion of what "modularity" is? Also, it seems to me that there is a natural connection between modularity as I understand it and "compositionality" as used in linguistics. Does anyone have any information, references, or ideas on these points? -Lee Altenberg ------------------------------ Date: Thu, 22 Aug 85 17:20 EST From: Clarke Thacher Subject: Expert Systems A professor in our Psychiatry department has expressed interest in any work which has been done with Expert Systems in psychiatry (please not ELIZA). He is interested in it as a diagnostic tool to be used by the physician (and for teaching third year medical students). Please send any leads to: Clarke Thacher BITNET: UKC323@UKCC Computing Center University Of Kentucky Lexington, Ky. ------------------------------ Date: Wed, 21 Aug 85 13:28:56 pdt From: aurora!eugene@RIACS.ARPA (Eugene miya) Subject: Additional comment about master bibliography Oh, I forgot one MAJOR point of maintenance work. I am just now receiving smaller bibiliographies on things like computer networks. There are many collisions with papers already in the existing file. The problem is subtle because of slight variations in annotation styles which bibliographic sorting programs cannot appropriately handle. Also, transferring interesting comments and annotations from one entry to another is also time consuming. Two smaller bibliographies have come from England, and differences in spelling are another subtle problem: Defense and Defence. --eugene miya NASA Ames Research Center {hplabs,hao,dual,ihnp4,vortex}!ames!aurora!eugene emiya@ames-vmsb.ARPA ------------------------------ Date: 20 Aug 1985 02:54:13-EDT, Tue, 20 Aug 85 02:37 EDT From: straz@AQUINAS.THINK.COM@MIT-CCC, Steve Strassmann Subject: Good news and bad news, Mr. Wizard... [Forwarded by BNevin@BBNCCH.] >From the September issue of "Science 85": "This is a good time to play a scientist on TV. Researchers at the University of Pennsylvania say that scientists on the tube are warm, attractive, and five times more likely to be virtuous than villainous. But the study also showed that TV scientists are killed more often than soldiers, private eyes, and police officers." ------------------------------ Date: 25 Aug 1985 10:25:51 PDT From: Stuart Cracraft Subject: Drew Liao's comments about chess "I too believe that a computer should learn how to play chess before it is allowed to play in a tournament rather than rely on moves encoded into the program." - Drew Liao, AILIST V3 #102, 1-Aug-85 The above doesn't make much sense to me. Currently, chess programs such as Belle and Cray Blitz usually play no more than the first 10 moves from a pre-stored "opening book". If the opponent makes an extremely odd or unusual move early, retrieval from the book is terminated and normal tree-searching is begun in order to generate a move. What would Drew have us do? Turn off book completely? Rely only on tree-search for the opening? The opening is extremely tricky, because pawn configurations and piece placements are being set up for 20 moves later. Thus, most programs that rely on heuristics and tree-search for opening play are prone to fall into traps a book usually avoids. They are also prone to jumble their pieces in bad ways. Therefore, I argue that allowing opening book is essential to good play. A more valid criticism would concern the transition from opening book to tree-searching. Many programs deal with this transition very poorly. International Masters or Grand- masters can often take advantage of their poor *TRANSITION* play and mop up quickly in a positional sense. I see no great difference between storing 50,000 opening positions in a computer "book" and a human expert spending 5 weeks studying the Caro-Kann defense. Both have memorized in order to avoid re-creating extensive work *OTHERS* have done for them ahead of time. Why re-invent the wheel? Stuart Cracraft ------------------------------ Date: Mon, 19 Aug 1985 13:30 EDT From: Hewitt@MIT-MC Subject: Prolog will fail as the foundation for AI Misconceptions? From: PEREIRA at SRI-AI.ARPA Carl Hewitt's message is based on several misconceptions: 1. (the least interesting one) All the so-called commercially viable Prolog systems in Lisp are not really Prolog systems written IN Lisp, but rather Prolog systems written FOR Lisp machines. Or is it that a microcode sublanguage or Lisp machine pointer-smashing operations are part of List as we know it? Yes. They are DEFINITELY part of Common Lisp as we know it being implementations of reading and writing operations on record structures. Such implementation methods are NOT part of Logic as a Programming language. Without those machine-level operations, those Prolog systems would run too slow and use too much memory to be useful for serious Prolog programming. From the Prolog implementation point of view, what is important about the Lisp machines is not that they run Lisp, but that they can be microcoded and have good support for tagged data types and stack operations. It is important to many users that they can make use of ALL the software available to the community and not just be limited to the tiny amount in Prolog. Furthermore in the future good software will be ported from stand alone Prolog systems to Prolog implemented on Lisp. However to good Lisp software will not be able to be ported to the stand alone Prolog systems. 2. If the decisions (actions) of a system are not determined by its inputs, the system is nondeterministic. Nondeterminism in a system can be either an artifact of our incomplete knowledge (or lack of interest) of the detailed operation of the system; or it can be ``real physical'' nondeterminism. It would take us to far to discuss whether the second kind of nondeterminism is ``real'' or also an artifact. In any case, most uses of nondeterminism, say in models of concurrency, are of the first kind, and can be expressed appropriately in various temporal/dynamic logics. Admittedly, these are not Prolog, but then Common Lisp is not Lisp 1.5! (Prolog is 13 years old, Lisp 25). Yes indeed there is a large problem here that poses fundamental problems for using Logic as a Programming language to make decisions in Open Systems. 3. The first logic course dictum ``from a contradiction one can conclude anything'' is getting in the way. Notice that the dictum says ``can'', not ``must''. There is an enormous difference between things that are in principle true and things that an agent knows to be true in a way that can affect its action. An agent might know ``p'' and ``not p'', but it might well never come to infer the dreaded totally unrelated ``q'' which IN PRINCIPLE follows from the contradiction. This inference might not happen either because of inference control mechanisms of the agent (eg. it uses the set-of-support strategy) or because the agent's logic is just TOO WEAK to conclude anything from a contradiction (vide Hector Levesque's work in the proceedings of the last AAAI). In any case, Horn clauses (the basis of Prolog) are too weak to represent contradictions... :-) I claim that in practice the contradictions lie close to the surface and occur in any nontrivial application. Thus the contradictions pose fundamental problems for using Logic as a Programming Language. 4. The question of whether ``taking action'' fits in a logic paradigm tends to be answered negatively after an hour's worth of consideration. If you persist for several years, though, this question becomes a source of insight on the relations between knowledge, state and action that is not available to those who started by dismissing the question after that initial hour. There is just too much work on logics of knowledge and action in AI and computer science for me to try to discuss it here. Some of this work has been applied to logic programming, either in the form of new logic programming languages based on temporal or dynamic logics or in representations of temporal reasoning and decision in, say, Prolog. I have been thinking about the problem for many years having designed Micro-Planner, the first "procedural embedding of logic" programming language in 1967. I claim that the problem of taking action poses fundamental problems for using Logic as a Programming language. 5. It is curious to see someone by implication defend Lisp as a language for expressing the taking of action! I claim that current Lisp systems are better than current Prolog systems for taking action because the only ways to take action in current Prolog systems are kludges. We know by now that the most difficult issue in ``reactive systems'' is not EXPRESSING action, but rather keeping it under control to prevent unwanted interactions. In this area, less is REALLY more, and highly complex languages like Lisp are just not suitable for the formal reasoning about programs that is needed to help us believe our programs do what we intend. To those who say ``...but we can keep to a carefully constrained subset of Lisp, use only message passing for interactions,...'' I will answer that the history of all large Lisp programs I know (and I think that is a representative sample) is littered with the brutalized corpses of constrained programming styles. Anyone who has looked at the current flavor mechanism in Zetalisp and its use in the window system will know what I mean... 5. Underlying Carl Hewitt's misconceptions is the old chestnut ``logic is static, systems are dynamic''. Note that the above quotation is NOT anything that I said. Any language, be it first-order logic or Lisp, is static; it is its USE which is dynamic (changes the state of communicating agents). A good analogy here is the use of differential equations to model dynamic systems in classical mechanics. The differential equations themselves do not say directly what happens when (they are ``static'' in Hewitt's jargon). I do not deny that dynamic systems can be DESCRIBED using logic only that they can be CONTROLLED. It is their SOLUTIONS that tell us the sequence of events. Even the solutions are given as static objects (functions from an open interval of the reals to some space). Does anyone worry that such equations do not ``really'' describe the dynamic behavior of a system? Is it not possible to combine such ``static'' entities with an incremental solution procedure to build systems that actually control their (classical mechanical) environment? I do not believe that the control system can be implemented using Logic as a Programming language ------------------------------ Date: Tue 20 Aug 85 21:44:22-PDT From: Ashok Subramanian Subject: Seminar - Partial Evaluation in Meta Interpreters (SU) Prof. Ehud Shapiro, of the Weizmann Institute of Science, will present a talk at 9 am on Monday, the 26th of August, in Margaret Jacks Hall room 352. The Magic of Partial Evaluation, or Meta Interpreters for Real Ehud Shapiro The Weizmann Institute of Science Enhanced meta-interpreters can implement sophisticated functions within complex software system. Examples are explanation facilities in expert systems, algorithmic debuggers in programming environments, and layers of protection and control in operating systems. However, the execution overhead of the added layer of interpretation is unacceptable in many applications. Partial evaluation can eliminate the overhead of meta-interpreters. A partial-evaluator can specialize an enhanced meta-interpreter with respect to a given program, generating a variant of this program which inherits the enhanced functionality of the meta-interpreter, but not its overhead. An application of a Concurrent Prolog partial evaluator to operating system development will be shown. ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Wed Aug 28 10:01:01 1985 Date: Wed, 28 Aug 85 10:00:56 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: RO Received: from sri-ai.arpa by csnet-relay.arpa id a006956; 26 Aug 85 16:20 EDT Date: Mon 26 Aug 1985 12:27-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #114 To: AIList@SRI-AI Received: from rand-relay by vpi; Wed, 28 Aug 85 09:40 EST AIList Digest Tuesday, 27 Aug 1985 Volume 3 : Issue 114 Today's Topics: Literature - AI in Business & Recent Technical Reports & Recent Articles ---------------------------------------------------------------------- Date: 20 Aug 1985 21:05-EST From: leff%smu.csnet@csnet-relay.arpa Subject: AI in Business Expert Systems, Artificial Intelligence in Business by Pual Harmon and David King is now available from Library of Computer and Information Science Publishers Price $24.95 Members Price $19.95 ------------------------------ Date: 20 Aug 1985 21:44-EST From: leff%smu.csnet@csnet-relay.arpa Subject: Recent Technical Reports Addresses for ordering same: Research Institute for Advanced Computer Science Mail STop 230-5, NASA/Ames Research Center Moffett Field, California 94035 ATTENTION: Technical Librarian Department of Computer Science 405 Upson Hall Cornell University Ithaca, New York 14853 Erna Amerman Department of Comptuer Science University of Illinois at Urbana-Champaign 1304 West Springfield Avenue Urbana, Illinois 61801 Technical Reports Computer Sciences Department University of Wisconsin 1210 West Dayton Street Madison, Wisconsin 53706 Ms. Brenda Ramsey UCLA Computer Science Department 3732 Boelter Hall Los Angeles, CA 90024 Carnegie Mellon University Computer Science Department Pittsburgh, PA 15213 ____ Bibliography Entries __________________________________________________ %A Jeffrey Alan Jackson %T Economics of Automatic Generation of Rules From Examples in A Chess End Game %R Department of Computer Science File No. 132 %I University of Illinois at Urbana-Champaign %D FEB 1985 %A Robert Stepp %T A Description and User's Guide for Cluster/2 A Program for Conjunctive Conceptual Clustering %R Department of Computer Science Report No. 1085 %I University of Illinois at Urbana-Champaign %D FEB 1985 %A G. Smolka %A P. Panangaden %T FRESH: A Higher-Order Language with Unification and Multiple Results %R 85-685 %I Department of Computer Science, Cornell University %C Ithaca, New York %D MAY 1985 %A Rick Briggs %T An Approach to Deeper Expert Systems %R 84.11 %I Research Institute for Advanced Computer Science, NASA/AMES Research Center %C Moffett Field, California %A Charles F. Neveu %A Charles R. Dyer %A Roland T. Chin %T Object Recognition Using Hough Pyramids %R TR 576 %I The University of Wisconsin-Madison Computer Sciences Department %D JAN 1985 %A Deborah A. Joseph %A W. Harry Plantinga %T On the Complexity of Reachability and Motion Planning Questions %R TR586 %I The University of Wisconsin-Madison Computer Science Department %D FEB 1985 %A Udi Manber %T A Distributed Implementation of Backtracking %R TR588 %I University of Wisconsin-Madison Department of Computer Sciences %D MAR 1985 %A Matthew R. Korn %A Charles R. Dyer %T 3-D Multiview Object Representations for Model-Based Object Recognition %R TR602 %I University of Wisconsin-Madison Department of Computer Sciences %D JUN 1985 %A Richard Preston Hooper %A Michel A. Melkanoff %T An Application of Knowledge-Based Systems to Electronic Computer-Aided Engineering, Design and Manufacturing Data Base Transport %R CSD-850011 %I Computer Science Department, UCLA %K IGES %X describes a method of developing methodology for transferring databases between CADCAM systems There is a charge of $19.25 for this item %A Randal E. Bryant %T Symbolic Verification of MOS Circuits %I Carnegie Mellon University Department of Computer Science %D APR 1985 %X The program MOSSYS simulates the behavior of a MOS circuit represented as a switch-level symbolically. That is, during simulator the user can set an input to either 0, 1 or a Boolean variable. The simulator then computes the behavior of hte circuit as a function of past and present and input variables. By using heuristically efficient Boolean function manipulation algorithms, the verification of a circuit by symbolic simualtion can proceed much more quickly than by exhaustive logic simulation. In this paper we present our concept of symbolic simualtion, dervie an algorithm for switch-level symbolic simulation, and present experimental measurements from MOSSYM %A Jon Doyle %T Reasoned Assumptions and Pareto Optimality %I Carnegie Mellon University Department of Computer Science %D JAN 1985 %A D. M. McKeown Jr. %A J. F. Pane %T Alignment and Connection of Fragmented Linear Features in Aerial Imagery %I Carnegie Mellon University Department of Computer Science %D APR 1985 %A Gary Kahn %A John McDermott %T MUD: A Drilling Fluids Consultant %I Carnegie Mellon University Department of Computer Science %D MAR 1985 %A Geoffrey E. Hinton %T Distributed Representation %I Carnegie Mellon University Department of Computer Science %D OCT 1984 %A Theodore F. Lehr %T The Implementation of a Production System Machine %I Carnegie Mellon University Department of Computer Science %D MAY 1985 %A Steven Linton %T A Game-Playing Porgram that Learns by Analyzing Examples %I Carnegie Mellon University Department of Computer Science %D MAY 1985 %A Jaime G. Carbonnel %T Derviational Analogy: A Theory of Reconstructive Problem Solving and Expertise Acquisition %I Carnegie Mellon University Department of Computer Science %D MAR 1985 ------------------------------ Date: 22 Aug 1985 02:23-EST From: leff%smu.csnet@csnet-relay.arpa Subject: Recent Articles %A M. A. Covington %T A Further Note on Looping in Prolog %J SIGPLAN %V 20 %N 8 %D AUG 1985 %P 28-31 %A D. Nute %T A Programming Solution to Certain Problems with Loops in Prolog %J SIGPLAN %V 20 %N 8 %D AUG 1985 %P 32-37 %A D. Poole %A R. Goebel %T On Eliminating Loops in Prolog %J SIGPLAN %V 20 %N 8 %D AUG 1985 %P 38-40 %A Mark Stefik %T Strategic Computing at DARPA: Overview and Assessment %J Communications of the ACM %D JUL 1985 %V 28 %N 7 %P 690-703 %X discusses various projects at DARPA. Here are the time schedules for various things they want as extracted from the Commerce Business Daily request for proposals: Autonomous Land Vehicle 1985 - The vehicle is expected to traverse a 20-km route on a paved road at up to 10km per hour. The vehicle will carry out only forward motion, without obstacle avoidance 1986 The vehicle is expected to maneuver to avoid small fixed polyhedral objects spaced 100 meters 1987 The vehicle will be able to plan and execute a route across 10km of open desert at speeds up to 5 km per hour. It should demonstrate an understanding of types of soil and ground cover. 1988 The vehicle should plan and execute a 20-km route on a road network, using landmarks as a navigation aid. To avoid obstacles, the vehicle will have to maneuver off the road. 1989 The vehicle should traverse across country at 10 km per hour avoiding obstacles. 1990 The route traversed by the vehicle will include wooded terrain, paved and unpaved roads, and deserts. The vehicle may have to consolidate multiple goals Pilots Associate (R2D2 for military aircraft) goals vague or unspecified Aircraft Carrier Battle-Management system phase 1 - look at military database and reason about ships and submarines, determine their readiness for missions and the effects of redirecting them phase 2 - handle five times real time performance and to achieve performance ten thousand times current performance phase 3 - provide aid to commanders in evaluating alternatives Expert Systems 1986 - Demonstrate capabilities for situation assessment where conclusions are annotated with different levels of confidence, support 3000 rule databases at 1000 inferences per second (1/3 of real time) 1989 - support speech input, increase speed factor of three 1992 - support multiple cooperating expert systems increase speed by factor of five Image Understanding 1986 - demonstrate image-understanding for vehicle on simple terrain 1988 - be able to recognize land marks 1990 - navigation on complex terrains 1992 - recognize targets and threats in battlefield Speech Production and Understanding 1986 100 words vocabulary, speaker dependent, sever noise 1988 1000 word, continuous speech, speech dependent, low noise 1990 200 word vocabulary, speaker independent, sever noise 1992 1000 word, continous speech, speaker independent, natural grammar They claim that computation will hit 40 million inferences per second for 1986 milestones and 20 billion for 1992 Natural language 1986 - demonstrate natural language interface for queries to database 1988 - should understand paragraphs about air threat 1990 - be able to converse and actively help user form a plan 1993 - interactive multiuser system and understand streams of information %A I. Peterson %T Conversing with Computers Naturally %J Science News %V 128 %D JUL 27, 1985 %N 4 %P 53 %K microcomputers natural language database Bozena H. Thompson Frederick B. Thompson Microrim Savvy Caltech Natural Access System %X discusses Intellect Microrim, and Savvy and a Natural Language System developed by Bozena H. Thompson and Frederick B. Thompson for IBM PC's %A Eric Nee %T Xerox to Transfer Some IS Operations to Shugart Plant %J Electronic News %D AUG 5, 1985 %V 31 %N 1561 %P 24 %X Xerox will transfer its Artificial Intelligence Business from Pasadena to Sunnyvale %T VLSI-Chip Test System Tests Itself at Board Level %J Electronics %D AUG 5, 1985 %P 46-49 %V 58 %N 31 %K MegaOne Expert-Like diagnostic %X Mega-One has introduced a VLSI test system that can diagnose malfunctions in itself. The manufacturer claims that it is "expert-like", i. e. is an expert system but is not set up using production rules %A Tom Manuel %T Tektronix Makes Major Commitment to AI Market %J Electronics %D AUG 5, 1985 %P 46-49 %V 58 %N 31 %X Tektronix is lowering its price on its 32032 based microcomputer that it is billing as an AI machines. It has added two more models, and added a 32-bit object oreinted machine. %A Adam B. Green %T Searching for Product X %J InfoWorld %D AUG 5, 1985 %P 28 %V 7 %N 31 %K prolog microcomputers %X Adam Green, noted for his pushing DBASE products, says that the new runaway product (like visicalc and database) will be the based on Prolog %T Updates %J Datamation %P 117 %D AUG 1, 1985 %V 31 %N 15 %K Gary Moskowitz, Xerox, natural language, office systems %X says that AI should be aim to proofread documents for grammar errors; and to help collaboration between humans %T VLSI Design System Uses Artificial Intelligence %J IEEE Computer Graphics and Applications %D AUG 1985 %P 89 %V 5 %N 8 %K Applicon BRAVO! design rule %X Applicon has introduced BRAVO!, a VLSI system which uses AI to monitor circuitry for design rule compliance and which will redesign layouts. %T IBM Adds 3 Programs for AI Applications %J Electronic News %D AUG 12, 1985 %V 31 %N 1562 %P 24 %K expert system tool lisp programming database %X IBM introduced VM Programming in Logic (A Prolog compiler). IBM provides communications with VM/SP, SQL/DS and LISP/VM. They also introduced Expert System Consulation environment and Expert System Development environment for building and using expert systems. %T Control Data, Digicon Sign 3-Year Value Added Agreement %J Electronics %V 31 %N 1562 %P 25 %K CDC lisp prolog expert system tool %X CDC introduced Lisp/VE, Prolog/VE, KES/VE %T Tektronix unwraps AI workstations, Lisp Version %J ComputerWorld %V 19 %N 32 %D AUG 12, 1985 %P 16 %X same info as other article on Tektronix above %T News, World Digest %J ComputerWorld %V 19 %N 32 %P 24 %K education Australia Queensland %X Queensland Secondary schools will introduce a new computer curriculum which includes AI. %T HP Gives 3.3 million for AI Research %J Electronics %D AUG 19, 1985 %P 23 %V 58 %N 33 %K University of Pennsylvania %X TI is giving some work stations to Pennsylvania. Appears to be part of a program of giving their work stations to schools announced elsewhere. %A Kevin Smith %T Britain Makes Major Bid to Build Commercial Fifth-Generation Machine %J Electronics %D JUL 8, 1985 %P 26-27 %V 58 %N 27 %K Alice Declarative Alvey Compiler Target Language Flagship HOPE ICL %X describes various parallel architectures being investigated by Britain. They hope to establish a defacto standard, beating IBM to the punch and to get a commercial product up. %A John Gallant %T AI Product Deluge hits DP Market %J ComputerWorld %V 19 %N 33 %P 1+ %D AUG 19, 1985 %K DM Data Howard Dicken DEC IBM LISP Expert System Tool Prolog CDC David Hertz %X IBM has announced Prolog and Expert System tools for it VM operating system. Charges: Prolog (Programming in Logic): $8,000 Expert System Consultation Environment/VM (delivery front end for expert system tool): $25,000 or a monthly charge of $1250.00 Expert System Development Environment (to make expert systems for use by above tool): $35,000 or a monthly charge of $1750.00. DEC has announced an AI VAXStation which is a MICROVAX II without a floppy drive and with various languages. This includes an implementation of Common Lisp, and they have upgraded OPS5 and have marketing agreements with Quintus for its Prolog and Artificial Intelligence Corporation for its Intellect front end for its database. Quintus Prolog will cost $6,000. CDC announcement of tools: PROLOG/VE (version of C-prolog): $4620 to $36330 LISP/VE (Common Lisp): $5166 to $40614 KES/VE (Knowledge Engineering System): $11,424 to $70,594 Also an interview with various people on the marketing impact of these announcements. Howard Dicken, publisher of AI trends, says this "will legitimize AI." However, other people feel be hurt. They draw parallel with the IBM personal computer which on one legitimized the microcomputer industry but also took away market share and eliminated some smaller companies. %T Random Access %J ComputerWorld %V 19 %N 33 %P 2 %D AUG 19, 1985 %K Intellicorp KEE simulation microcomputer expert system %X Intellicorp will issue the third release of their Knowledge Engineering Kit. They also announced Simkit which will be used to create knowledge-based simulation software and PC-Host which is an implementation of the KEE system on "conventional architecture computers." %A Howard Morgan %T The Microcomputer and Decision Support %J ComputerWorld %V 19 %N 33 %P 39-46 %D AUG 19, 1985 %K Lightyear microcomputer Expert-Ease %X discusses the use of AI in MIS, particularly Light-Year and Expert Systems in MIS on page 44 of this article. %A Mitch Betta %T AI Specialist Sees $5 billion expert systems mart by 1990 %J ComputerWorld %V 19 %N 33 %P 84 %D AUG 19, 1985 %K marketing Social Security Administration Sperry Corpration Atle Fjeld %X Atle Fjeld gave a briefing in which he discussed AI. He gave the example of automating Social Security eligibility rules as an example of its use. Sperry anticipates spending 200 million dollars over the next five years and hopes to capture 15 to 20 percent of the market. The Sperry Corporation Knowledge System Center employs 150 people. %T TI Acquires Ten Percent of Carnegie Group %J Electronic News %D AUG 19, 1985 %V 31 %N 1563 %P 8 %K Explorer Knowledge Craft Language Craft %X TI will fund research at the Carnegie Group and receive an internal license for use of its products. Carnegie Group is expected to install 20 of TI's explorer systems over the next 18 months. %A Michael Bunken %T Dec Enters Artificial Intelligence Market with Workstation %J Electronic News %D AUG 19, 1985 %V 31 %N 1563 %K Vaxstation Lisp Prolog Quintas OPS-5 Common Lisp %X same info on DEC as in article in Computerworld above %A Nicholas Ourusoff %T The Physical Symbol System Hypothesis of Newell and Simon: A Classroom Demonstration of Artificial Intelligence %J SIGCSE Bulletin %P 19-23 %V 17 %N 3 %D SEP 1985 ------------------------------ End of AIList Digest ******************** From comsat@vpics1 Thu Aug 29 00:06:01 1985 Date: Thu, 29 Aug 85 00:05:50 edt From: comsat@vpics1.VPI To: fox@opus (FRANCE,RDJ,JOSLIN,ROACH,FOX) Subject: From: AIList Moderator Kenneth Laws Status: R Received: from sri-ai.arpa by csnet-relay.arpa id a022041; 28 Aug 85 13:52 EDT Date: Wed 28 Aug 1985 09:12-PDT Reply-to: AIList@SRI-AI US-Mail: SRI Int., 333 Ravenswood Ave., Menlo Park, CA 94025 Phone: (415) 859-6467 Subject: AIList Digest V3 #115 To: AIList@SRI-AI Received: from rand-relay by vpi; Wed, 28 Aug 85 23:48 EST AIList Digest Wednesday, 28 Aug 1985 Volume 3 : Issue 115 Today's Topics: Seminar - Mathematical Learning (MIT), Conferences - Symposium on Factory Automation and Robotics & Simulation, Automation, and Robotics & Program of Expert Systems in Government Symposium ---------------------------------------------------------------------- Date: Mon, 26 Aug 85 09:59:20 EDT From: Andy diSessa Subject: Seminar - Mathematical Learning (MIT) [Forwarded from the MIT bboard by SASW@MIT-MC.] Marlene Kliman will talk about her work on balancing this Weds., August 28, 4 PM, third floor conference room. Mathematical Knowledge Underlying Performance on the Balance Scale Interviews of elementary school children using a balance scale show that balancing is a complex domain involving many levels of competence and understanding. Performance on set balancing tasks may not always be indicative of underlying knowledge, as children frequently display special-case knowledge. As children experiment with the balance scale, they not only find new pieces of knowledge, but they also begin to find connections among pieces of knowledge. This talk will present an analysis of mathematical knowledge underlying performance on the balance scale in terms of the kinds of things children do and generalizations they make when exploring the balance scale, strategies children find natural and helpful, and facets of balancing that children find particularly intriguing. Educational implications of this analysis will be discussed. ------------------------------ Date: 27 Aug 85 10:33 EDT From: (Herb Bernstein) Subject: Symposium on Fact. Aut. & Robotics Reminder: The Symposium on Factory Automation and Robotics A Forum for Industrial and Academic Robotics Engineers and Scientists will be held 9-11 September 1985 by the Courant Institute of Mathematical Sciences, NYU in honor of Marvin Denicoff sponsored by the National Science Foundation Registration $35 ($25 for NYU faculty and staff), in advance to NYU/CIMS Symposium on Fac. Aut. and Robotics Courant Institute of Mathematical Sciences 251 Mercer Street, New York, N.Y. 10012 Attn: Herbert J. Bernstein Or at the meeting, Eisner and Lubin Auditorium, Loeb Student Center, 566 LaGuardia Place (corner of LaGuardia Place and Wash. Sq. South). For more information or to RSVP, mail to yaya@nyu (on ARPANET), or call 212-533-3363 or 212-460-7444. ------------------------------ Date: 20 Aug 1985 21:09-EST From: leff%smu.csnet@csnet-relay.arpa Subject: Conference Announcement I am relaying this from printed material received here for the benefit of those who might be interested. I am not connected with the conference. Society of Computer Simulation P. O. Box 17900 San Diego California 92117-7900 Telephone 619-277-3888 Southwestern Region Simualtion presents their Fall 85 Technical Meeting Theme: Simulation and Automation and Robotics Dates: October 24-25 1985 Thursday and Friday Location: Fort Worth Texas Cost: $25 registration fee Keynote Address: Robotics and Intelligent Systems: An Overview by George A. Bekey ------------------------------ Date: 28 Aug 85 08:21:59 EDT (Wed) From: Marshall D. Abrams Subject: Program of Expert Systems in Government Symposium Following is a list of all the sessions and papers for the Expert Systems in Government Symposium to be held October 24-25, 1985 at the Tysons Westpark Hotel, McLean, VA. The symposium is sponsored by the MITRE Corporation and the IEEE Computer Society. Two one-day tutorials are scheduled for October 23rd: "Concepts of Knowledge Engineering" by Kamran Parsaye and "Expert Systems Development" by Elaine Kant. Additional information is available from abrams@mitre. A registration form follows the session listing. ENGINEERING APPLICATIONS Vice-Chairman: Prof. Mary Lou Maher, Carnegie-Mellon University Session-Chairman: Dr. Duvvuru Sriram, Carnegie-Mellon University "A Rule-Based System for Masonry Failures," S.M. Cornick, Carlton University "Development of An Intelligent Interface To An Interactive Design Model," R.A. Harris, Vanderbilt University "Micro Computer Based Expert Systems In Engineering: An Exam- ple," Nitin Pandit, D. Sriram, Carnegie-Mellon University "LASER: A Programming Environment For Building High Performance Expert Systems," Dr. Y.V. Reddy, West Virginia University "An Expert Tutor for Rigid Body Mechanics: Athena Cats-- Macavity," John H. Slater, Robert B.P. Petrossiam, S. Shyam- sunder, Massachusetts Institute of Technology MISSION PLANNING Vice_Chairman: Prof. Mark Fox, Carnegie-Mellon University Sessions_Chairman: Dr. Robert Milne, Pentagon "Management of AI System Software Development For Military Deci- sion Aids," Kerry Gates, John Lemmer, PAR Technology Corporation "Expert Mission Planning and Replanning Scheduling System," G.B. Hankins, J.W. Jordan, J.L. Katz, A.M. Mulvehill, The MITRE Corporation "Mission Planning Within the Framework of the Blackboard Model,"Glen Pearson, FMC Corporation "Developing a Microcomputer Based Intelligent Project Planning System," Suzanne Bradley, Ruth Buys, Amr ElSawy, Alan Sipes, The MITRE Corporation PANEL: FRONTIERS OF KBES: PRO & CON MODERATOR: Dr. Marvin Denicoff, Thinking Machine, Inc. Panelists: PRO Prof. Chandrasekaran, Ohio State University PRO Dr. Neil Pundit, DEC CON Dr. Gary Martins, Intelligent S/W, Inc. CON Dr. John Benoit, MITRE MIDDLE Dr. Robert Milne, Pentagon MIDDLE Dr. Kamran Parsaye, Intelliware UNCERTAINTY MANAGEMENT Vice_Chairman: Dr. North Fowler, RADC Sessions_Chairman: Dr. Piero Bonissone, G.E. "An Application of Fuzzy Reasoning to A Confirmation/Disconfirmation Decision Making Algorithm," Joseph A. Karakowski, U.S. Army "An Expert System Based on a Stochastic Parallel Network," V. Venkatasubramanian, Northeastern University "Data Analysis Assisted by an Uncertainty Varying Expert System," Petr Hajek, Fred Neil Springsteel "In Introduction to Fact Based Model Expert Systems," Sing Chi Koo, Software Intelligence Lab, Inc. NETWORK MANAGEMENT Vice_Chairman: Dr. Richard Martin, MCC Session_Chairman: Dr. Greg Vesonder, AT&T Bell Laboratories "Specifications of a Knowledge System for Packet-Switched Data Network Topological Design," Chelsea White, III, Edward A. Sykes, University of Virginia "Telecommunications Resource Allocation A Knowledge-Based Sys- tem," Dai Chuang, Booz Allen & Hamilton "Compass: An Expert System for Telephone Switch Maintenance," Shri K. Goyal, David S. Prerau, Alan V. Lemmon, Alan Gunderson, Robert Reinke, GTE Laboratories, Inc. "NEMESYS: An Expert System for Fighting Congestion in the At&T Network," Dr. Richard C. Windecker, Stephen M. Guattery, Dr. Francisco J. Villarreal, AT&T Bell Labs, Inc. INTELLIGENT TUTORING SYSTEM Vice_Chairman: Prof Saj-Nicole Joni, Yale Session_Chairman: Prof. Beverly Woolf, University of Mas- sachusetts "Intelligent Tutoring Using the Integrated Diagnostic Model: An Expert System for Diagnosis and Repair," Howard R. Smith, Pamela K. Fink, John C. Lusth, Southwest Research Institute "An Informal Programming Language," Jeffery Bonar, William Weil, University of Pittsburg "Understanding Discourse Conventions in Tutoring," Beverly Woolf, University of Massachusetts "PRUST: An Automatic System Program Debugger," Lewis Johnson, ISI NEW DIRECTIONS IN KNOWLEDGE ACQUISITION Vice_Chairman: Dr. Bruce Bullock, Teknowledge Session_Chairman: Ethan Scarl, The MITRE Corporation "Artificial Intelligence for C:3:I: The Design and Development of a Prototype Causal Schema Knowledge-Based System," Gerard W. Hopple, Stanley M. Halpin, Defense Systems, Inc. "Capturing Domain Primitives for Knowledge Engineering," R. Peter Bonasso, The MITRE Corporation "A Shared Knowledge Base for Independent Problem Solving Agents," Susan E. Conry, R.A. Meyer, J.E. Searleman, Clarkson College "Research in the Applications of Expert Systems at the NASA Ames-Dryden Flight Research Facility," Eugene L. Duke, Victoria A. Regeniek, NASA Ames Research Center INTELLIGENT ANALYSIS Vice_Chairman: Dr. Morten Hirschberg, Army Ballistic Research Laboratories Session_Chairman: Major Russ Frew, U.S. Army Intelligence Center & School "XCOR--A Knowledge-Based System for Correction of Oceanographic Reports," Vincent G. Sigillito, R.F. Wachter, The Johns Hopkins University "A Demonstration of An Ocean Surveillance Information Fusion Ex- pert System," Elizabeth H. Groundwater, Science Applications Int., Corporation "An AI Technology Insertion Experiment With Analyst," Peter Bonasso, The MITRE Corporation "ACES Airborne Communication Expert System: A Proposed Expert Sys- tem for Managing Airborne Military Communication," Hal Miller- Jacobs, The MITRE Corporation "Artificial Intelligence Applications to Intelligence Analysis," Jerald L. Feinstein, Booz Allen & Hamilton, Inc. PANEL: INTELLIGENT COMPUTER ASSISTED INSTRUCTION MODERATOR: Dr. Jeff Bonar, LRDC/University of Pittsburg Panelists: 1. Dr. Beverly Woolf, University of Massachusetts. 2. Dr. Joseph Psotka, PERI-IC, Army Research Institute 3. Prof Walker Schneider, University of Illinois 4. Prof. T. Govindraj, Georgia Instute of Technology 5. Dr. Stelan Ohlsson, University of Pittsburg SOFTWARE Vice_Chairman: Dr. Santosh Chokhani, The MITRE Corporation Session_Chairman: Dr. Robert Ensor, AT&T Bell Laboratories "Automation of Programming: the ISFI Experiments," Richard Brown, The MITRE Corporation "Mutations & Their Consequences; A Study of Non-Monotonic Behavior," Gary A. Cleveland, Richard Brown, The MITRE Corpora- tion "Designing Expert Systems for Ease of Change," Judith N. Frosch- er, Robert J.K. Jacob, Department of the Navy "Arrowsmith-P A Prototype Expert System for Software Engineering Management,: Victor R. Basili, Connie Loggia Ramsey, University of Maryland LANGUAGE TOOLS AND TECHNIQUES Vice_Chairman: Dr. Thomas London, AT&T Bell Laboratories Session_Chairman: Dr. Kamran Parsaye, Intelliware, Inc. "A Multicriteria Model to Select An Expert System Shell," Ernest H. Forman, Thomas J. Nagy, The George Washington University "Guess/1: A General Purpose Expert Systems Shell," Newton S. Lee, John W. Roach, Virginia Polytechnic Institute and State University "Prolog for Expert Systems: An Evaluation," Richard Helm, Cath- erine Lassez, Kim Marriott, University of Melbourne "Evaluating the Existing Tools for Developing Expert Systems In PC Environment," Animesh Karna and Amitabh Karna , Information Technical Institutes WEAPONS SYSTEMS: ADAPTIVE CONTROL Vice_Chairman: Dr. Morton Hirschberg, Army Ballistics Research Laboratories Session_Chairman: Dr. Morton Hirschberg, Army Ballistics Research Laboratories "A Numerical Symbolic Expert System in Computational Fluid Dynam- ics," Rick Briggs, NASA Ames Research Center "RICA: An Expert System for Radar Image Classification," Deborah T. Franks, Software Architecture & Engineering, Inc. "Concepts for a Tactical Fire Control Decision Aid," Richard C. Kaste, U.S. Army Ballistic Research Laboratory "An Experimental Expert Weapon Direction System," R.L. Stewart, D.R. Ousborne, The Johns Hopkins University PANEL: NETWORK MANAGEMENT MODERATOR: Dr. Kamal Karna, The MITRE Corporation Panelists: 1. Dr. Shri Goyal, GTE Labs 2. Dr. Richard Wolf, AT&T Bell Labs 3. Ms. Alice Van Domelen, GTE Sprint 4. Mr. Lucien Capone, Jr., Vice President, Booz Allen & Hamil- ton SYSTEMS ENGINEERING Vice_Chairman: Ms. Diane Tosh, E Systems, Melpar Division Session_Chairman: Ray Dolgert, Softtech, Inc. "An Expert System Prototype For Aiding in the Development of Software Functional Requirements For NASA Goddard's Command Management System," Jay Liebowitz, The George Washington Univer- sity "Comments on the Procurement & Development of Expert Systems," David L. Hall, HRB Singer "Distributed Intelligence In Alternative Analysis for Computor Systems Selection & Configuration," Fiorin T. Zeviar, Boeing Com- puter Services Co. "Expert Systems & Robotics for the Space Station: Design Con- siderations," Barry G. Silverman, V.S. Moustakis, The George Washington University INTELLIGENT INFORMATION RETRIEVAL-1 Vice_Chairman: Dr. Roy Rada, National Library of Medicine Session_Chairman: Dr. Brian McCune, Advanced Information & Deci- sion Systems "A Rule-Oriented Methodology for Constructing a Knowledge Base from Natural Language Documents," Dr. Stan Matwin, University of Ottawa "Understanding Technical Documents and Graphics," R.P. Futrelle, University of Illinois "Information Retrieval Techniques In An Expert System for Foster Care," Sheila G. Winett, Edward A. Fox, Virginia Polytechnic In- stitute "Cognitive Graphing & The Representation of Biomedical Knowledge," G. Matthew Bonham, George J. Nozicka, F.N. Stokman, The American University ELECTRONIC WARFARE Vice_Chairman: Dr. Morton Hirschberg, Army Ballistic Research Laboratories Sessions_Chairman: Mr. Joseph Mitola, Advanced Information & De- cision Systems "Using Artificial Intelligence Techniques in a War Gaming En- vironment," Gregory B. White, Stephen E. Cross, Air Force Insti- tute of Technology "Evidential Reasoning for Electronic Warfare Threat Assessment," Dr. Tom Garvey, SRI International "Object Oriented Modeling in Electronic Warfare, Jim Cunningham, Advanced Information & Decision Systems "Information Management Expert Systems, Capt Robert Milne, Penta- gon DESIGN, MONITORING & CONTROL Vice_Chairman: Dr. Eric Braude, RCA Session_Chairman: Dr. Peter Politakis, DEC "A Knowledge-Based System for Transit Bus Maintenance," Peter Wood, The MITRE Corporation "Reactor Safety Assessment System--A Situation Assessment Aid for USNRC Emergency Response," Michael A. Bray, Idaho National En- gineering Laboratory "Artificial Intelligence System for Failure Detection & Monitor- ing in Electronics & Communication," L.F. Pau, Battelle Insti- tute, Geneva Laboratories "TITAN: An Expert System to Assist in Troubleshooting the Texas Instruments 990 Minicomputer System," Joe D. Stuart, Steven D. Pardue, Radian Corporation INTELLIGENT INFORMATION RETRIEVAL-2 Vice_Chairman: Dr. Roy Rada, National Library of Medicine Session_Chairman: Dr. Roy Rada, National Library of Medicine "An Expert System for Document Retrieval," Bruce Croft, Roger H. Thompson, University of Massachusetts "Building A Knowledge Base Statistically," Hafedh Mili, Roy Rada, George Washington University "The Application of an Expert System for Information Retrieval at the National Archives," Daniel DeSalvo, Jay Liebowitz, American Management Systems, Inc. "Intelligent Information Retrieval: Issues in User Modelling," Robert R. Korfhage, So. Methodist University MEDICINE-1 Vice_Chairman: Prof. B. Chandrasekaran, Session_Chairmen: Dr. R. Smith, Ohio State University "A Deductive Inference & Associative Memory Retrieval," James A. Reggia, University of Maryland "An Expert Advisory System for Primary Eye Care in Developing Countries," Jack H.. Ostroff, Chandler R. Dawson, John K. Kastner, Sholom Weiss, Casimir A. Kulikowski, Kevin B. Kern, Rutgers University "Knowledge Representation for Knowledge Directed Data Retrieval," Jon Sticklen, Ohio State University "Mapping Medical Knowledge Into the Conceptual Structure Representation Language," Thomas C. Bylander, Ohio State Univer- sity FLEXIBLE AUTOMATION Vice_Chairman: Prof. Tulin Mangir, UCLA Session_Chairman: Roger Duncan, The MITRE Corporation "The Evolution of an Expert System for Process Planning, Roy S. Freedman, Hazeltine Corporation "Flexible Automation for Printed Circuit Board Assembly, Dr. Robert J. Stewart, Westinghouse Manufacturing Systems and Tech- nology Center "Sensor-Based Robot Programming for Automated Assembly," Dr. Shaheen Ahmed, Purdue University MEDICINE-2 Vice_Chairman: Prof. Sholom Weiss, Rutgers University Session_Chairmen: Dr. R. Smith, Ohio State University "RED: A Classification & Abductive Inference Expert System," Jack W. Smith, Jr., Ohio State University "Improved Retrieval Through Traversal of a Knowledge-Base," Ellen Bicknell-Brown, Roy Rada, National Institutes of Health "Semantic Network Representations for Neurological Diagnosis," S. Srihari, Xiang, Chatkow, Shapiro, SUNY at Buffalo "An Expert System for Interpretation of Cranial C.T. Scan Im- ages," R. Kremar, S. Srihari, SUNY at Buffalo ENVIRONMENT & WEATHER Vice_Chairman: Dr. Robert Kay, National Oceanic & Atmospheric Administration Session_Chairman: Dr. Jude Franklin, Planning Research Corpora- tion "Expert Systems for Environmental Regulation," Susan G. Hadden, Chandler Stolp, University of Texas at Austin "A Demonstration Expert System For Weather Forecasting," George Swetnam, The MITRE Corporation "An Expert System for Water Quality Protection Permits," Charles Spooner, U.S. Environmental Protection Agency "The Potential Role of Artificial Intelligence/Expert Systems in the Warning and Forecast Operations of the National Weather Ser- vice," Randy I. Racer, John Gaffney, Jr., National Weather Ser- vice INTELLIGENT INTERFACES Vice_Chairman: Prof. Saj-Nicole Joni, Yale University Session_Chairman: Prof. Beverly Woolf, University of Mas- sachusetts "A Knowledge Based Lab Assistant for a Computer Based Instruction System," Larry Christensen, Gordon Stokes, Bill Hays, E. Dale Coons, Brigham Young University "The Cognitive Principles Underlying Software Design," B. Adel- son, Yale University "Tutoring Expertise: Human & Otherwise," David Littman, Jeannine Pinto, Elliot Soloway, Yale University "Application of Expert Systems to Training," David L. Young, Mys- tech Associates, Inc. POLITICS & WELFARE Vice_Chairman: Prof. Sholom Weiss, Rutgers University Session_Chairman: Prof. Gavan Duffy, University of Texas at Aus- tin, AMI Micro System "A Prolog Model of Social Structure," Sanjoy Banerjee, Baruch College "WEDS Welfare Eligibility Determination System: An Expert System in an Administrative Context," Eswaran Subrahmanian, Carnegie- Mellon University "Expert Systems as Elite Foreign Policy Advisors: Some User/Machine and Organization/Machine Issues," Howard Tamashiro, G. Brunk, University of Oklahoma "A Role for Expert System in Foreign Policy Decision Making," Prof. Stewart Thorson, Dr. Christie Anderson, Syracuse Universi- ty "SYSFIL: A Generalized Expert System Architecture for Filings," Krishan Chhabra, Kamal Karna, The MITRE Corporation This list of session was compiled 8/20/85 and is subject to further correction. send REGISTRATION to: Expert Systems Symposium IEEE Computer Society 1730 Massachusetts Avenue, NW Washington, DC 20036-1903 prior to 10/1/85 late reg. surcharge member non-member tutorial $200 $250 $30 symposium $120 $160 $30 Give membership #, name, title, company, address, etc. Blood type is optional (joke). On-site registration may be limited. Call 202/371-0101 for latest information on space availability. ------------------------------ End of AIList Digest ********************