Can a niching method locate multiple attractors embedded in the hopfield network?

Akira Imada, Keijiro Araki

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

We apply evolutionary computations to the Hopfield’s nenral network model of associative memory. In the model, a number of patterns can be stored in the network as attractors if synaptic weights are determined appropriately. So far, we have explored weight space to search for the optimal weight configuration that creates attractors at the location of patterns to be stored. In this paper, on the other hand, we explore pattern space to search for attractors that are created by a fixed weight configuration. All the solutions in this case are a priori known. The purpose of this paper is to study the ability of a niching genetic algorithm to locate these multiple solutions using the Hopfield model as a test function.

Original languageEnglish
Title of host publicationSimulated Evolution and Learning - 2nd Asia-Pacific Conference on Simulated Evolution and Learning, SEAL 1998, Selected Papers
EditorsBob McKay, Xin Yao, Charles S. Newton, Jong-Hwan Kim, Takeshi Furuhashi
PublisherSpringer Verlag
Pages325-332
Number of pages8
ISBN (Print)3540659072, 9783540659075
DOIs
Publication statusPublished - 1999
Event2nd Asia-Pacific Conference on Simulated Evolution and Learning, SEAL 1998 - Canberra, Australia
Duration: Nov 24 1998Nov 27 1998

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume1585
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other2nd Asia-Pacific Conference on Simulated Evolution and Learning, SEAL 1998
Country/TerritoryAustralia
CityCanberra
Period11/24/9811/27/98

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint

Dive into the research topics of 'Can a niching method locate multiple attractors embedded in the hopfield network?'. Together they form a unique fingerprint.

Cite this