This report stems from an April 1994 meeting, organized by AAAI at the suggestion of Steve Cross and Gio Wiederhold.1 The purpose of the meeting was to assist ARPA in defining an agenda for foundational AI research. Prior to the meeting, the fellows and officers of AAAI, as well as the report committee members, were asked to recommend areas in which major research thrusts could yield significant scientific gain -- with high potential impact on DOD applications -- over the next ten years. At the meeting, these suggestions and their relevance to current national needs and challenges in computing were discussed and debated. An initial draft of this report was circulated to the fellows and officers. The final report has benefited greatly from their comments and from textual revisions contributed by Joseph Halpern, Fernando Pereira, and Dana Nau.

Efficient indexing schemes have influenced the acceptance of production systems in the industrial world. However, in embedded-control systems, production systems have not been applied intensively because of their nondeterministic run-time behavior. Thus, nonpredictability of response times is a major obstacle to the widespread use of expert systems in the real-time domain. The RETE and TREAT algorithms and their offspring play a major role in the implementation of efficient pattern-matching systems. Therefore, it is worthwhile to investigate run-time predictability for these match algorithms. This article presents three different schemes for estimating the time needed for operations in the production-system execution model.

Given a knowledge base KB containing first-order and statistical facts, we consider a principled method, called the random-worlds method, for computing a degree of belief that some formula Phi holds given KB. If we are reasoning about a world or system consisting of N individuals, then we can consider all possible worlds, or first-order models, withdomain {1,...,N} that satisfy KB, and compute thefraction of them in which Phi is true. We define the degree of belief to be the asymptotic value of this fraction as N grows large. We show that when the vocabulary underlying Phi andKB uses constants and unary predicates only, we can naturally associate an entropy with each world. As N grows larger,there are many more worlds with higher entropy. Therefore, we can usea maximum-entropy computation to compute the degree of belief. This result is in a similar spirit to previous work in physics and artificial intelligence, but is far more general. Of equal interest to the result itself are the limitations on its scope. Most importantly, the restriction to unary predicates seems necessary. Although the random-worlds method makes sense in general, the connection to maximum entropy seems to disappear in the non-unary case. These observations suggest unexpected limitations to the applicability of maximum-entropy methods.

Information extraction is the task of automaticallypicking up information of interest from an unconstrained text. Informationof interest is usually extracted in two steps. First, sentence level processing locates relevant pieces of information scatteredthroughout the text; second, discourse processing merges coreferential information to generate the output. In the first step, pieces of information are locally identified without recognizing any relationships among them. A key word search or simple patternsearch can achieve this purpose. The second step requires deeperknowledge in order to understand relationships among separately identified pieces of information. Previous information extraction systems focused on the first step, partly because they were not required to link up each piece of information with other pieces. To link the extracted pieces of information and map them onto a structuredoutput format, complex discourse processing is essential. This paperreports on a Japanese information extraction system that merges information using a pattern matcher and discourse processor. Evaluationresults show a high level of system performance which approaches human performance.

This article describes a new system for induction of oblique decision trees. This system, OC1, combines deterministic hill-climbing with two forms of randomization to find a good oblique split (in the form of a hyperplane) at each node of a decision tree. Oblique decision tree methods are tuned especially for domains in which the attributes are numeric, although they can be adapted to symbolic or mixed symbolic/numeric attributes. We present extensive empirical studies, using both real and artificial data, that analyze OC1's ability to construct oblique trees that are smaller and more accurate than their axis-parallel counterparts. We also examine the benefits of randomization for the construction of oblique decision trees.

The human orientation system is a complex system in which the brain merges information from a variety of sensors to help maintain a coherent interpretation of body position and movement. I designed a model of this system based on the observer theory model (OTM), which was developed by Merfeld (1990) for the orientation system of the squirrel monkey. Under this scheme, the central nervous system has an internal representation of the sensor organs and tries to minimize the error between its estimate of the sensory afferent signals and the actual afferent signals. It works iteratively until the results of the proposed experiment can be modeled.

Recent general AI conferences show a decline in both the number and the quality of vision papers, but there is tremendous growth in, and specialization of, computer vision conferences. Hence, one might conclude that computer vision is parting or has parted company with AI. This article proposes that the divorce of computer vision and AI suggested here is actually an open marriage: Although computer vision is developing through its own research agenda, there are many shared areas of interest, and many of the key goals, assumptions, and characteristics of computer vision are also clearly found in AI.

Representatives of universities and businesses were chosen by the Japan Technology Evaluation Center to investigate the state of the technology in Japan relative to the United States. The panel's report focused on applications, tools, and research and development in universities and industry and on major national projects.

Real-time problem solving is not only reasoning about time, it is also reasoning in time. Many techniques, mostly ad hoc, have been developed in both the real-time community and the AI community for solving problems within time constraints. This article is an attempt to step back from the details and examine the entire issue of real-time problem solving from first principles. We examine the degrees of freedom available in structuring the problem space and the search process to reduce problem-solving variations and produce satisficing solutions within the time available.

The Seventh International Workshop on Qualitative Reasoning about Physical Systems was held on 16-19 May 1993. The bulk of the 50 attendees work in the AI area, but several engineers and cognitive psychologists also attended. The two topics attracting special attention were automated modeling and the design task. This article briefly describes some of the presentations and discussions held during the workshop.