Bayesian inference nets 190 algorithms, structure-moving task 209-17 Kononenko's treatment 201-2 ALG(SIG, E) 50

They produce one set of rules from one set of data and have no memory which permits them to add to a knowledge base by further learning. Incremental learning systems remember the concepts which they have learned and can use them for further learning and problem solving. Some examples are, CONFUCIUS (Cohen 1978) and Marvin (Sammut 1981). These programs build a model of their task environment through successive learning experiences which require interaction with the environment. The task that we consider in this paper involves a program learning to control an agent in a reactive environment. This is an environment where changes occur in response to actions. Agents other than the learner may be present. As an agent accumulates experience, it constructs a world model or theory of behaviour which can be used to predict the outcome f Present address: Department of Computer Science, University of New South Wales, Sydney, Australia.

In this paper we will be concerned with such reasoning in its most general form, that is, in inferences that are defeasible: given more information, we may retract them. The purpose of this paper is to introduce a form of non-monotonic inference based on the notion of a partial model of the world. We take partial models to reflect our partial knowledge of the true state of affairs. We then define non-monotonic inference as the process of filling in unknown parts of the model with conjectures: statements that could turn out to be false, given more complete knowledge. To take a standard example from default reasoning: since most birds can fly, if Tweety is a bird it is reasonable to assume that she can fly, at least in the absence of any information to the contrary. We thus have some justification for filling in our partial picture of the world with this conjecture. If our knowledge includes the fact that Tweety is an ostrich, then no such justification exists, and the conjecture must be retracted.

The particular task in question is the interpretation of the electrical signals generated by the heart muscle, known as the electrocardiographic (Eco) interpretation.

This paper describes some work on automatically generating finite counterexamples in topology, and the use of counterexamples to speed up proof discovery in intermediate analysis, and gives some examples theorems where human provers are aided in proof discovery by the use of examples.

R1, a knowledge-based configurer of VAX-11 computer systems, began to be used over a year ago by Digital Equipment Corporation's manufacturing organization. The success of this program and the existence at DEC of a newly formed group capable of supporting knowledge-based programs has led other groups at DEC to support the development of programs that can be used in conjunction with RI. This paper describes XSEL, a program being developed at Carnegie-Mellon University that will assist salespeople in tailoring computer systems to fit the needs of customers. XSEL will have two kinds of expertise: it will know how to select hardware and software components that fulfil the requirements of particular sets of applications, and it will know how to provide satisfying explanations in the computer system sales domain.

The equivalence problem for program schemes, or for programs, is reduced to the proving of a theorem in second-order logic. This work extends Manna's first-order logic reductions. Some examples of the technique are given together with a suggested method for obtaining proofs in special cases by firstorder methods. INTRODUCTION Several workers in recent years have considered using techniques and ideas of various mathematical theories of computation for proving interesting results about computer programs. This paper is concerned with two of these approaches.

Using the Neomycin rule base, and inputting Neomycin's own actions to the action justification generator, the average size of J(.4,) was ten and the maximum size was approximately one hundred. When an Odysseus-generated rule base for the Neomycin domain was used, these set sizes increased by a factor of four to five. After the set J(Ai) is generated, the action justification ranking subsystem of Odysseus determines the likelihood that J(Ai) contains ji, the action justification of the specialist. This involves, first, ranking ji,„ in order of likelihood of being equal to the unknown An example of ranking rule is: given two elements of a J(.4,), where,4, occurs early in the problem solving session, the

Some of the more popular alternativo used to build knowledge systems are production systems, backward-chained reasoning, logic programming, heuristic search, and the Blackboard framework. Many of the applications implemented in production systems have been written in the OPS language [8]. In this framework, knowledge is represented as a set of homogeneous rules that are scanned for applicability in a data base that contains the current state of solution. Backward chaining also has a homogeneous set of rules, but the search for applicable rules is driven by a hierarchy of goals and sub-goals. The best known system for implementing this type of program is EMYCIN [4].

The impact of a computer-based data management system on the completeness of clinical trial data was studied before and after the system's introduction in an oncology clinic. Physicians use the system, termed ONCOCIN, to record data during patient visits and to receive advice about treatment and tests required by experimental cancer protocols. Although ONCOCIN does not force the user to enter all data expected by the protocol, after its introduction there was improvement in the recording frequency of such data. The percentage of expected physical findings recorded increased from 74% to 91% (p .05),