Evolving multilayer neural networks using permutation free encoding technique

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

Abstract

In this paper we present an evolutionary system using genetic algorithm (GA) for evolving artificial neural networks (ANNs). Existing genetic algorithms for evolving ANNs suffer from the permutation problem as a result of recombination. Here we propose a novel encoding scheme for representing ANNs which avoids the permutation problem while efficiently evolving multilayer ANN architectures. The evolutionary system has been implemented and tested on a number of benchmark problems in machine learning and neural networks. Experimental results suggest that the system shows superiority in performance, in most of the cases.

Original languageEnglish (US)
Title of host publicationProceedings of the 2009 International Conference on Artificial Intelligence, ICAI 2009
Pages32-38
Number of pages7
StatePublished - 2009
Event2009 International Conference on Artificial Intelligence, ICAI 2009 - Las Vegas, NV, United States
Duration: Jul 13 2009Jul 16 2009

Publication series

NameProceedings of the 2009 International Conference on Artificial Intelligence, ICAI 2009
Volume1

Other

Other2009 International Conference on Artificial Intelligence, ICAI 2009
Country/TerritoryUnited States
CityLas Vegas, NV
Period7/13/097/16/09

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'Evolving multilayer neural networks using permutation free encoding technique'. Together they form a unique fingerprint.

Cite this