A Two-Level Hybrid Architecture for Structuring Knowledge for Commonsense Reasoning
Published 18 August 2007
Ron Sun
Citations8
Generate an AI Snapshot to get a quick, structured summary of this paper.
Study Snapshot
ObjectiveStudy objective
MethodsResearch methodology
PopulationPopulation studied
Sample sizeSample sizes
OutcomesStudy outcomes here
ResultsStudy results comes here
LimitationsResearch study limitations comes here
A concise AI-generated summary of the paper will appear here once you click Generate AI Snapshot.
TL;DR
In this chapter, a connectionist architecture for structuring knowledge in vague and continuous domains is proposed, and it consists of an inference network with nodes representing concepts and links representing rules connecting concepts and a microfeature-based replica of the first level.
Abstract
In this chapter, a connectionist architecture for structuring knowledge in vague and continuous domains is proposed. The architecture is hybrid in terms of representation, and it consists of two levels: one is an inference network with nodes representing concepts and links representing rules connecting concepts, and the other is a microfeature-based replica of the first level. Based on the interaction between the concept nodes and microfeature nodes in the architecture, inferences are facilitated and knowledge not explicitly encoded in a system can be deduced via a mixture of similarity matching and rule application. The architecture is able to take account of many important desiderata of plausible commonsense reasoning, and produces sensible conclusions.
Keywords
Computer Science
Neural NetworksMultilayer feedforward networks are universal approximators
21,248 Citations1989Kurt Hornik, Maxwell B. Stinchcombe +1 more
It is rigorously established that standard multilayer feedforward networks with as few as one hidden layer using arbitrary squashing functions are capable of approximating any Borel measurable function from one finite dimensional space to another to any desired degree of accuracy, provided sufficiently many hidden units are available.
Psychological ReviewFeatures of similarity.
7,287 Citations1977Amos Tversky
The metric and dimensional assumptions that underlie the geometric representation of similarity are questioned on both theoretical and empirical grounds and a set of qualitative assumptions are shown to imply the contrast model, which expresses the similarity between objects as a linear combination of the measures of their common and distinctive features.
Cambridge University Press eBooksSimilarity and Analogical Reasoning
2,397 Citations1989Stella Vosniadou, Stella Vosniadou +22 more
Zeitschrift für wirtschaftlichen FabrikbetriebBuilding Expert Systems
1,949 Citations1986
This book provides a broad introduction to the concepts and methods necessary for an understanding of how these systems work.
Elsevier eBooksSymbolic Logic and Mechanical Theorem Proving
1,472 Citations1973
This book contains an introduction to symbolic logic and a thorough discussion of mechanical theorem proving its applications and how it can be applied to various areas such as question answering, problem solving, program analysis, and program synthesis.
Psychology Press eBooksInside Case-Based Reasoning
1,409 Citations2013Christopher K. Riesbeck, Roger C. Schank
CBR tends to be a good approach for rich, complex domains in which there are myriad ways to generalize a case, and is similar to the rule-induction algorithms of machine learning.
Language<b>Parallel Distributed Processing:</b> Explorations in the microstructures of cognition. By David E. Rumelhart, James L. McClelland, and the PDP Research Group. Vol. I, Foundations; Vol. II, Psychological and biological models. Cambridge, MA: MIT Press, 1986. Pp. xx, 547; xii, 611.
741 Citations1987Geoffrey Sampson
Communications of the ACMConnectionist expert systems
665 Citations1988Stephen I. Gallant
Connectionist networks can be used as expert system knowledge bases and can be constructed from training examples by machine learning techniques, giving a way to automate the generation of expert systems for classification problems.
Cognitive ScienceMassively Parallel Parsing: A Strongly Interactive Model of Natural Language Interpretation*
432 Citations1985David L. Waltz, Jordan Pollack
This work describes a parallel model for the representation of context and of the priming of concepts in a natural language processing system with modular knowledge sources but strongly interactive processing.
HAL (Le Centre pour la Communication Scientifique Directe)Non-standard logics for automated reasoning
406 Citations1988Philippe Smets
Refinement of approximate domain theories by knowledge-based neural networks
383 Citations1990Geoffrey G. Towell, Jude Shavlik +1 more
Cognitive ScienceThe Logic of Plausible Reasoning: A Core Theory
361 Citations1989Allan Collins, Ryszard S. Michalski
This is one of the first attempts to construct a formal theory that addresses both the semantic and parametric aspects of the kind of everyday reasoning that pervades.
HAL (Le Centre pour la Communication Scientifique Directe)An introduction to possibilistic and fuzzy logics
160 Citations1988Didier Dubois, Henri Prade +2 more
IEEE Transactions on Neural NetworksBack-propagation learning in expert networks
129 Citations1992R. C. Lacher, S.I. Hruska +1 more
The authors develop back-propagation learning for acyclic, event-driven networks in general and derive a specific algorithm for learning in EMYCIN-derived expert networks, which offers automation of the knowledge acquisition task for certainty factors, often the most difficult part of knowledge extraction.
Munich Personal RePEc Archive (Ludwig Maximilian University of Munich)Neural Representation of Conceptual Knowledge.
102 Citations1986Jerome A. Feldman
There is now good reason to reject both extreme views, but a compact encoding that derives from the punctate model appears to fit well with all the facts.
Psychology Press eBooksintegrating Marker Passing and Problem Solving
41 Citations2014James Hendler
It is described how the technique known as "marker-passing," a parallel, nondeductive, activation-spreading algorithm, can be used to enhance the choice mechanism in a problem-solving system.
Medical Entomology and ZoologyMundane reasoning by parallel constraint satisfaction
34 Citations1990Mark Derthick, Geoffrey E. Hinton
This thesis describes a frame system similar to KL-ONE, called micro-KLONE, for representing and reasoning about knowledge which may be incomplete or inconsistent, based on probabilistic sampling to find a single plausible model of the domain in order to answer a query.
Knowledge AcquisitionA connectionist model for commonsense reasoning incorporating rules and similarities
34 Citations1992Ron Sun
This work shows that connectionist models of reasoning are not just “implementations” of their symbolic counterparts, but better computational models of commonsense reasoning.
Cambridge University Press eBooksTwo-tiered concept meaning, inferential matching, and conceptual cohesiveness
32 Citations1989Ryszard S. Michalski
International Joint Conference on Artificial IntelligenceIntegrating knowledge-based system and neural network techniques for robotic skill acquisition
17 Citations1989David A. Handelman, Stephen H. Lane +1 more
An approach to robotic control that is patterned after models of human skill acquisition is described to develop robots capable of learning how to accomplish complex tasks using designer-supplied instructions and self induced practice.
International Joint Conference on Artificial IntelligenceMarker-passing and microfeatures
15 Citations1987James Hendler
This paper describes a mechanism which allows the two processes to be merged via the introduction into the symbolic network of a defining-characteristic link which is affects the spread of the symbolic information in a manner resembling local connectionist computations.
NatureMathematics of Inheritance
14 Citations1933Roger Fisher
Harry H. Laughlin remarks that the majority of characters in which practical men are interested are too complex genetically for their inheritance to have been analysed in terms of genes.
Information SciencesBeyond associative memories: Logics and variables in connectionist models
12 Citations1993Ron Sun
It is shown that there is a connection between propositional logics and the weighted-sum computation customarily used in connectionist models, which can handle Horn clause logic and Shoham's logic as special cases.
Recognition of semantically incorrect rules: a neural-network approach
11 Citations1990LiMin Fu
A novel technique that applies the neural-network learning strategy of back-propagation to recognize semantically incorrect rules is presented and the viability of this technique has been demonstrated in a practical domain.