Showing papers in "Artificial Intelligence in 1990"

TL;DR: In this article, the authors explore principles governing the rational balance among an agent's beliefs, goals, actions, and intentions, and show how agents can avoid intending all the foreseen side-effects of what they actually intend.

TL;DR: In this article, it was shown that probabilistic inference using belief networks is NP-hard and that it seems unlikely that an exact algorithm can be developed to perform inference efficiently over all classes of belief networks and that research should be directed toward the design of efficient special-case, average-case and approximation algorithms.

TL;DR: In this paper, a number of families of nonmonotonic consequence relations, defined in the style of Gentzen [13], are studied from both proof-theoretic and semantic points of view.

TL;DR: A variation of minimax lookahead search, and an analog to alpha-beta pruning that significantly improves the efficiency of the algorithm, and a new algorithm, called Real-Time-A∗, for interleaving planning and execution, which proves that the algorithm makes locally optimal decisions and is guaranteed to find a solution.

TL;DR: The tensor product representation rests on a principled analysis of structure; it saturates gracefully as larger structures are represented; it permits recursive construction of complex representations from simpler ones; it extends naturally to continuous structures and continuous representational patterns.

TL;DR: This paper presents a connectionist architecture which automatically develops compact distributed representations for variable-sized recursive data structures, as well as efficient accessing mechanisms for them.

TL;DR: An integrated strategy is described which utilizes the distinct advantages of each scheme and shows that, in hard problems, the average improvement realized by the integrated scheme is 20–25% higher than any of the individual schemes.

TL;DR: The ACT theory of skill acquisition and its PUPS successor provide production-system models of the acquisition of skills such as LISP programming, geometry theorm-proving, and solving of algebraic equations.

TL;DR: In this paper, the authors consider two approaches to giving semantics to first-order logics of probability, and provide axiom systems that are sound and complete in cases where a complete axiomatization is possible, showing that they do capture a great deal of interesting reasoning about probability.

TL;DR: It is suggested that a program which can demonstrate expertise in the area of medical consultation will have to use a judicious combination of categorical and probabilistic reasoning—the former to establish a sufficiently narrow context and the latter to make comparisons among hypotheses and eventually to recommend therapy.

TL;DR: This research attempts to show how some of these patterns of reasoning can be captured using only the classical notions of logic, by extending a logic of belief so that it is possible to say that only a certain proposition is believed.

TL;DR: This work abstracts from the usual numeric representations by encoding only qualitative relationships, which are inequality constraints on the joint probability distribution over the variables, and defines two types of qualitative relationship: Qualitative influences and Qualitative synergies.

TL;DR: This work has developed a multi-level theory of spatial representation of the environment based upon the observation and re-acquisition of distinctive visual events, i.e., landmarks, that provides the theoretical foundations for a visual memory database that smoothly integrates available metric knowledge of relative or absolute angles and distances.

TL;DR: ARS is a program that demonstrates how a set of semantic, structural, and pragmatic constraints can be used to select relevant analogs by forming a network of hypotheses and attempting to satisfy the constraints simultaneously.

TL;DR: Three different ways of mapping part-whole hierarchies into connectionist networks are described, suggesting that neural networks have two quite different methods for performing inference.

TL;DR: In pilot trials, the learning environment successfully taught novices to troubleshoot and to mentally simulate circuit behavior, and the implications of this work for the design of intelligent learning environments are explored.

TL;DR: The PRODIGY system as mentioned in this paper evaluates the costs and benefits of the control knowledge it learns and produces useful control knowledge by actively searching for "good" explanations that can be profitably employed to control problem solving.

TL;DR: A parallel distributed processing model is described that learns to comprehend single clause sentences that assigns thematic roles to sentence constituents, disambiguates ambiguous words, instantiates vague words, and elaborates implied roles.

TL;DR: Subsumption in terminologies turns out to be co-NP-complete for a minimal terminological representation language that is a subset of every useful terminological language.

TL;DR: This paper describes a successful approach to concept learning for heuristic classification that has been applied to the domain of clinical audiology and achieved a competence level equaling that of human experts and far surpassing that of other machine learning programs.

TL;DR: In this article, a new approach to inheritance reasoning in semantic networks allowing for multiple inheritance with exceptions is described, which leads to an analysis of defeasible inheritance which is both well-defined and intuitively attractive: it yields unambiguous results applied to any acyclic semantic network, and the results conform to our intuitions in those cases in which the intuitions themselves are firm and unambiguous.

TL;DR: In this paper, the authors present an approach to repair in which plan failures are described in terms of causal explanations of why they occurred, which are used to access abstract repair strategies, which can then be used to make specific changes to the faulty plans.

TL;DR: In this article, a series of revisions to Allen's theory of time and action is proposed in order to accommodate the possibility of continuous change, and it is argued that Allen's category of processes is superfluous, since it can be assimilated with the category of properties.

TL;DR: Two insights that arise when one studies the question "why and how is it that the authors say the same thing differently to different people, or even to the same person in different circumstances" are discussed.

TL;DR: It is proved that the problem solved by discrete relaxation (arc consistency) is log-space complete for P, which implies that discrete relaxation is inherently sequential and it is unlikely that it can solve the polynomial-time version of the consistent labeling problem in logarithmic time by using only aPolynomial number of processors.

TL;DR: Using the probably approximately correct framework developed in [12], Baum and Haussler have shown that if a neural network can be trained to automatically construct its own internal representations, then it might be better to settle for the system that works best.

TL;DR: This approach can be generalized if the set of components can be partitioned such that each of the components of a partition fail with equal probability but are much more or less likely to fail than those of other partitions.

TL;DR: A coherent way of interpolating three-dimensional data obtained by stereo, for example, with a simplicial polyhedral surface based on the use of the constrained Delaunay triangulation is proposed.

TL;DR: This paper describes a modification of brute-force search, singular extensions, that allows extensions to be identified in a dynamic, domain-independent, low-overhead manner.