A fault tolerant connectionist model for predicate logic reasoning, variable binding: using coarse-coded distributed representations

Authors:
Sriram G. Sanjeevi;Pushpak Bhattacharya
Affiliations:
Dept. of Comp. Science & Engg., N.I.T. Warangal, Warangal, India;Dept. of Comp. Science & Engg., I.I.T. Bombay, Mumbai, India
Venue:
NN'05 Proceedings of the 6th WSEAS international conference on Neural networks
Year:
2005

Citing 5
Cited 0

Distributed representations

Parallel distributed processing: explorations in the microstructure of cognition, vol. 1
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Connectionist inference models

Neural Networks
Advances in SHRUTI—A Neurally Motivated Model of RelationalKnowledge Representation and Rapid Inference Using Temporal Synchrony

Applied Intelligence
Artificial Intelligence: A Modern Approach

Artificial Intelligence: A Modern Approach

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, we describe a fault-tolerant model for reasoning using forward chaining for predicate logic rules and facts with coarse-coded distributed representations of instantiated predicates in a connectionist frame work. Distributed representations are known to give advantages of fault tolerance and graceful degradation of performance under noise conditions. The system supports usage of complex rules which involve multiple conjunctions and disjunctions. The system solves the variable binding problem in a novel way using coarse-coded distributed representations of instantiated predicates without the need to decode them into localist representations. System's performance with regard to its ability to exhibit fault tolerance under noise conditions is studied. The system offers better results of fault tolerance under noise conditions as compared to a connectionist reasoning system which uses localist representations. It has also exhibited better fault tolerance as compared to a Hopfield net for error correcting tasks of same magnitude.