Gul A. Agha's Actors: A Model of Concurrent Computation in Distributed Systems (The MIT Press, Cambridge, Mass., 1987, 144 pages, $25.00, ISBN O-262 010925) is part of the MIT Press Series in Artificial Intelligence. This volume is edited by Patrick Winston, Michael Brady, and Daniel Bobrow. In the actor formalism, pioneered by Carl Hewitt (1977), one perceives abstract computational agents, called actors, that are distributed in space. Each actor has a mailbox (and a mail address) and associated with each actor is a behavior. One actor can influence the actions of another actor only by sending it a communication.

Furthermore, the main conceptual foundations of AI--namely, the knowledge representation hypothesis of Brian Smith (1982) and the physical symbol system hypothesis of Allen Newell (1980)--are not discussed at all. These hypotheses have been considered fundamental cornerstones of AI research, but they are now being questioned as posing strong limitations on AI (Dahlbäck 1989; Dreyfus 1972; Winograd and Flores 1986). Given this perspective, the author concludes that AI's essential methodology is a continuous attempt to overcome the formal constraints of computer science and philosophy without sacrificing rigor. Although I liked the author's perspective, and I wholly agree with his main conclusion, both are just stated in the preface, and no further reference to them is given. Let's get a feeling of what this first volume is really about.

Li is not small compared to that of A. However, To understand how this rule works, let us return to the submarine example and assume that there are two groups of experts El,..., As is pointed out in Zadeh (1979a), the Dempster rule P*(notA) 1. This, in a nutshell, is the basic idea underly-of combination of evidence may lead to counterintuitive coning the Dempster-Shafer theory. The An important observation is in order at this juncture. P(A), that S is in A, the answer would be (after the object under consideration does not exist. P*(A) are the degrees of belief and plausibility associated of evidence, consider the following situation.

Philip Swarm Images and Understanding: Thoughts about Images, Ideas about Understanding, H. Barlow, C. Blakemore, and M. Weston-Smith, eds., A collection of essays based on a Rank Prize Fund's International Symposium, organized with the help of Jonathan Miller and held at the Royal Society in October 1986, Cambridge University Press, Cambridge, United Kingdom, 1990, 401 pp., ISBN O-521-34177-9 (cloth), ISBN O-521-36944-4 (paper). This volume is a well-written, informative, and thought-provoking collection of essays that should interest anyone concerned with the psychology of vision and visual communication. The aim of the original symposium was to bring together people from the arts and sciences who could present different perspectives on the subject of images and understanding. The result is an informal tour conducted by leading specialists (predominantly British) that visits both famous scientific battlefields and quaint artistic backwaters. Numerous striking pictures enliven the book: Here you can find the sensory somatic cortex of a bat, the British miners' leader Arthur Scargill in full rant, a notation for ballet, a mole used to advertise British Gas, instructions for righting a caravan, and many others.

You wait forever for one to come along, and then two come along at once. In this case, there has been a large gap in the market for a theoretical introduction to constraint programming ever since Edward Tsang's Foundations of Constraint Satisfaction (1993) went out of print. Therefore, we are very pleased to see two books written by two of the leading researchers in this field come along to fill the gap. Constraint programming is a very active research area within AI. It is a highly successful technology for solving a wide range of combinatorial problems, including scheduling, rostering, assignment, routing, and design. A number of companies, like ILOG, Dash Optimization, and Parc Technologies, market model building and constraint programming toolkits, which are used by companies as diverse as Amazon.com, Constraint programming is a declarative style of modeling combinatorial problems in which the user identifies the decision variables, their possible domain of values, and specifies constraints over the allowed values (for example, no two of these variables can take the same value). Sophisticated but general purpose AI search techniques like constraint propagation (to prune irrelevant parts of the search tree) and dependency directed backtracking can then be used to find solutions. Given the many advances made in constraint programming over the last decade, a new text would have been needed even if Edward Tsang's book had remained in print. These two new texts are written by two of the leading researchers in this field. Principles of Constraint Programming by Krzysztof Apt contains chapters that cover topics like local consistency, constraint propagation, linear equations, interval reasoning, and search. Constraint Processing by Rina Dechter covers similar ground but also has chapters that cover topics like local search, tree decomposition methods, optimization, and probabilistic networks more extensively. Dechter's book also contains a chapter by David Cohen and Peter Jeavons on tractability and one by Francesca Rossi on constraint logic programming. There is much in common between the two books. This is perhaps not so surprising since Krzysztof Apt thanks Rina Dechter for much useful discussion that helped him enter the field and start doing research in the area.

Because it assumes so much previous knowledge, the book will not be useful to the casual reader. One would be at a disadvantage without a reasonable familiarity with predicate calculus and modal logic, AI planning formalisms, and the work of Perrault and Allen on interpreting speech acts (for example, Allen and Perrault [1980]; Perrault and Allen [1980]). Accordingly, the reader of this review should be warned that my point of view is that of a researcher (specifically, an academic researcher) rather than a system builder; your mileage might vary. No review of this book would be complete without some mention of the commentaries, critical pieces written by other workshop participants that follow groups of related papers. Each commentator did an excellent job.

Stephen Grossberg The expanded edition of Perceptrons (MIT Press, Cambridge, Mass, 1988, 292 pp, $12.50) by Marvin L. Minsky and Seymour A. Papert comes at a time of unprecedented interest in the biological and technological modeling of neural networks. The one-year-old International Neural Network Society (INNS) already has over 3500 members from 38 countries and 49 U.S. states, with members joining at the rate of more than 200 per month. The American Association for Artificial Intelligence was, in fact, a cooperating society at the INNS First Annual Meeting in Boston on 6-10 September 1988. Hardly a week goes by in which a scientific meeting or special journal issue does not feature recent neural network research. Thus, substantive technical reviews or informed general assessments of the broad sweep of neural network research are most welcome to help interested scientists find their way into this rapidly evolving technology.

The author points out that "the ascription schema constrains mental ascriptions once a system is specified; but it puts no limit on which systems should have mental states ascribed to them." He postulates a "Supertrap" which strikes matches in the presence of gassoaked mice, topples dictionaries on mice, and, of course, snaps shut whenever mice nibble its bait "These habits betray a common malevolent thread, which is generalizable by (and only by) ascribing a persistent goal: dead mice." When we see other Supertrap behaviors, such as failure to harm cats that reek of gasoline, we become involved in a "semantic intrigue," an effort to understand how mental ascriptions cohere and interact. Whimsical examples aside, ascription is important for AI because it provides one more way to detect patterns that might otherwise go unnoticed. The ascription schema is proposed during the author's discussion of people's pragmatic sense.

CHINOOK that I also highly recommend. AI Magazine Volume 20 Number 1 (1999) ( AAAI) vided more than a glimpse of the intense process it described. One Jump Ahead was written by the person most involved in the process. Thus, it provides us with a direct view of Schaeffer's maturation--a maturation that we should all hope to have. Schaeffer does not pull any punches in his book; we see many of his elations, his disappointments, and his flaws.

Regardless of training, most people who come in contact with the field of AI are at least partially motivated by the glimmer of hope that they will get a better understanding of the mind. This quest, of course, is a rich and complex one. It is easy to get mired in minutiae along the way, be they the optimization of an algorithm, the details of a mental model, or the intricacies of a logical argument. Thagard's book attempts to call us back to the larger picture and to draw in new devotees--and, in general, he succeeds. This book begins, "Cognitive science is the interdisciplinary study of mind and intelligence..." (p.