Ray the adventurer was always eager to try new ideas and directions. He was not afraid to enter murky areas, and he always left them better illuminated. He introduced terms to the AI community such as default logic, closed-world assumption, and cognitive robotics; he opened avenues of theoretical research with new resolution proof methods and logics for nonmonotonic reasoning, diagnosis, and action; and he was the prime mover in the Cognitive Robotics initiative that has led to a whole new program of research. And he was an adventurer in more than just ideas. He frequently traveled to remote locations to add to his extraordinary collection of rare and exotic lepidoptera.

An informal workshop on concurrent logic programming, metaprogramming, and open systems was held at Xerox Palo Alto Research Center (PARC) on 8-9 September 1987 with support from the American Association for Artificial Intelligence. The 50 workshop participants came from the Japanese Fifth Generation Project (ICOT), the Weizmann Institute of Science in Israel, Imperial College in London, the Swedish Institute of Computer Science, Stanford University, the Massachusetts Institute of Technology (MIT), Carnegie-Mellon University (CMU), Cal Tech, Science University of Tokyo, Melbourne University, Calgary University, University of Wisconsin, Case Western Reserve, University of Oregon, Korea Advanced Institute of Science and Technology (KAIST), Quintus, Symbolics, IBM, and Xerox PARC. No proceedings were generated; instead, participants distributed copies of drafts, slides, and recent papers. A shared vision emerged from the morning session with concurrent logic programming fulfilling the same role that C and Assembler do now. Languages such as Flat Concurrent Prolog and Guarded Horn Clauses are seen as general-purpose, parallel machine languages and interface languages between hardware and software and not, as a newcomer to this field might expect, as high-level, AI, problemsolving languages.

ATP systems are used in a wide variety of domains: A mathematician might use the axioms of group theory to prove the conjecture that groups of order two are commutative; a management consultant might formulate axioms that describe how organizations grow and interact and, from these axioms, prove that organizational death rates decrease with age; or a frustrated teenager might formulate the jumbled faces of a Rubik's cube as a conjecture and prove, from axioms that describe legal changes to the cube's configuration, that the cube can be rearranged to the solution state. All these tasks can be performed by an ATP system, given an appropriate formulation of the problem as axioms, hypotheses, and a conjecture. Most commonly, ATP systems are embedded as components of larger, more complex software systems, and in this context, the ATP systems are required to autonomously solve subproblems that are generated by the overall system. To build a useful ATP system, several issues have to ...

In this article, the body of a report on automated deduction is presented that notes some significant achievements and takes a studied look at the future of the field. Mechanization of the deductive process includes not only proving new mathematical results by computer but also formally verifying the correctness of (certain properties of) computer chip designs and programs and even deducing the programs themselves from formal specifications of the task. A less obvious application is the use of automated inference tools within programming languages and within programs that produce scheduling algorithms and optimize other programs. Some of the early work in this field is already part of the fabric of the AI world. The machinery developed in this field is useful in inferring missing rules or facts in a problem specification (abduction) and generalizing from examples to full specifications (inductive inference, learning from examples), although I do not deal with these forms of reasoning in this article.

The eight sections are (1) Artificial Intelligence (introductory material); (2) Problem-Solving (search and game playing); (3) Knowledge and Reasoning (propositional and predicate logic, inference techniques, knowledge representation); (4) Acting Logically (planning); (5) Uncertain Knowledge and Reasoning (probabilistic reasoning, Bayesian nets, decision-theoretic techniques); (6) Learning (inductive learning, neural nets, reinforcement learning); (7) Communicating, Perceiving, and Acting (natural language processing, computer vision, robotics); and (8) Conclusions (philosophical foundations and summary). What makes this textbook so good? First, it is remarkably comprehensive. In the preface, the authors suggest several alternative paths through the book that could serve as the basis of a one-semester course. At the University of Pittsburgh, my colleagues and I cover roughly the first half of the book (Sections 1-4) in the firstsemester introductory graduate AI course, covering most of Sections 5 through 8 in a second-semester course.

One of the fastest advancing areas of modern science is functional genomics. This science seeks to understand how the complete complement of molecular components of living organisms (nucleic acid, protein, small molecules, and so on) interact together to form living organisms. Functional genomics is of interest to AI because the relationship between machines and living organisms is central to AI and because the field is an instructive and fun domain to apply and sharpen AI tools and ideas, requiring complex knowledge representation, reasoning, learning, and so on. This article describes two machine learning (inductive logic programming [ILP])-based approaches to the bioinformatic problem of predicting protein function from amino acid sequence. The first approach is based on using ILP as a way of bootstrapping from conventional sequence-based homology methods.

Recent developments have clarified the process of generating partially ordered, partially specified sequences of actions whose execution will achieve an agent's goal. Thus, it should be no surprise that the quest of building intelligent agents has forced AI researchers to investigate algorithms for generating appropriate actions in a timely fashion. Of course, the problem is not yet solved, but considerable progress has been made. In particular, AI researchers have developed two complementary approaches to the problem of generating these actions: (1) planning and (2) situated action. These two techniques have different strengths and weaknesses, as I illustrate later.

The American Association for Artificial Intelligence held its 1993 Fall Symposium Series on October 22-24 in Raleigh, North Carolina. This article contains summaries of the five symposia that were conducted: Automated Deduction in Nonstandard Logics; Games: Planning and Learning; Human-Computer Collaboration: Reconciling Theory, Synthesizing Practice; Instantiating Intelligent Agents; and Machine Learning and Computer Vision: What, Why, and How? This article contains summaries of the five symposia that were conducted: Automated Deduction in Nonstandard Logics; Games: Planning and Learning; Human-Computer Collaboration: Reconciling Theory, Synthesizing Practice; Instantiating Intelligent Agents; and Machine Learning and Computer Vision: What, Why, and How? Over the past decade, a wide variety of methods have been developed for automating deduction, with an even wider variety of nonstandard logics. The goals of the Automated Deduction in Nonstandard Logics Symposium were to bring together researchers working in this area with the aim of promoting comparisons of the various deduction methods that have been proposed, foster discussion of the different problems involved in automating the various logics, and obtain reports on the use of theorem provers for nonstandard logics in AI systems.

In October 1981,.Japan announced a national project to develop highly innovative computer systems for the 199Os, with the title "Fifth Generation Computer Systems " This paper is a personal view of that project, its significance, and reactions to it. The informatiou available to me at the time of writing (June 1982) is not as complete as I would have liked, a.nd Prcscnted at the Pergarnon Infotcch State of the Art Conference on "Japan and the Fifth Generation " in London, England, 27-29 Septemher 1982 Technical Note 265 I apologise for any mistakes or misinterpretations I may therefore have made. ETL and, in particular, to Kazuhiro includes "expert systems" and natural language interfaces. Thus VLSI technology is to be exploited to build advanced parallel architectures for AItype applications, where the basic machine language will be an extension of the logic programming language Prolog. So logic programming and Prolog play a crucial role in the systems envisaged.

We have developed a sales-support system for door-to-door sales of cosmetics based on a system called Skin-Expert, a skin-image grading service that includes analysis and diagnosis. Several parameters are extracted by image processing, and the skin grading is done by rules generated by data mining from a baseline of grades given by human skincare experts. Communication with the Skin-Expert is through a cellular telephone with a camera, using email software and a Web browser. Salespeople photograph the customer's skin using the camera in a standard cellular telephone and then send an email message that includes the picture as an attachment to our analysis system. Other parameters associated with the customer (for example, age and gender) are included in the body of the message.