An important limitation of traditional logic programming as a knowledge representation tool, in comparison with classical logic, is that logic programming does not allow us to deal directly with incomplete information. In order to overcome this limitation, we extend the class of general logic programs by including classical negation, in addition to negation-as-failure. The semantics of such extended programs is based on the method of stable models. The concept of a disjunctive database can be extended in a similar way. We show that some facts of commonsense knowledge can be represented by logic programs and disjunctive databases more easily when classical negation is available.

The Robotics Institute, Carnegie Mellon University Pittsburgh, PA 15213 Abstract In non-cooperative multi-agent planning, resolution of multiple conflicting goals is the result of finding compromise solutions. Previous research has dealt with such multi-agent problems where planning goals are well-specified, subgoals can be enumerated, and the utilities associated with subgoals known. Our research extends the domain of problems to include non-cooperative multi-agent interactions where planning goals are ill-specified, subgoals cannot be enumerated, and the associated utilities are not precisely known. Negotiation is performed through proposal and modification of goal relaxations. Case-Based Reasoning is integrated with the use of multi-attribute utilities to portray tradeoffs and propose novel goal relaxations and compromises. Persuasive arguments are generated and used as a mechanism to dynamically change the agents' utilities so that convergence to an acceptable compromise can be achieved.

In Proceedings of the Fifth International Conference on Machine Learning, pages 54-64, Ann Arbor, Michigan.

This paper describes a Bayesian technique for unsupervised classification of data and its computer implementation, AutoClass. Given real valued or discrete data, AutoClass determines the most probable number of classes present in the data, the most probable descriptions of those classes, and each object's probability of membership in each class. The program performs as well as or better than other automatic classification systems when run on the same data and contains no ad hoc similarity measures or stopping criteria. AutoClass has been applied to several databases in which it has discovered classes representing previously unsuspected phenomena.

Did you know your Organization can subscribe to the ACM Digital Library?

This article introduces a class of incremental learning procedures specializedfor prediction that is, for using past experience with an incompletely knownsystem to predict its future behavior. Whereas conventional prediction-learningmethods assign credit by means of the difference between predicted and actual outcomes,tile new methods assign credit by means of the difference between temporallysuccessive predictions. Although such temporal-difference method~ have been used inSamuel's checker player, Holland's bucket brigade, and the author's Adaptive HeuristicCritic, they have remained poorly understood. Here we prove their convergenceand optimality for special cases and relate them to supervised-learning methods. Formost real-world prediction problems, telnporal-differenee methods require less memoryand less peak computation than conventional methods and they produce moreaccurate predictions. We argue that most problems to which supervised learningis currently applied are really prediction problemsMachine Learning 3: 9-44, erratum p. 377

"This is a popular interpretation of expert systems - computer models providing artificial intelligence to managers and businessmen - which are transforming business in the 1980s. The book gives an inside view of many international companies such as Canon, Westinghouse, IBM, Du Pont and others." New York: Times Books.

"Expert systems are among the most exciting computer applications to emerge in the last decade. They allow a computer program to use expertise to assist in a variety of problems, such as diagnosing equipment failures and designing new equipment. Utilizing the results of artificial intelligence (AI) work on problem solving, they have become a commercially successful demonstration of the power of AI techniques. Correspondingly, by testing current AI methods in applied contexts, expert systems provide important feedback to the science about the strengths and limitations of these methods. In this review, we present the fundamental considerations in constructing an expert system, assess the state of the art, and indicate directions for future research. Our discussion focuses on the computer science issues, as opposed to issues of management or application." Annual Review of Computer Science.

Did you know your Organization can subscribe to the ACM Digital Library?

A set of hypotheses is formulated for a connectionist approach to cognitive modeling.