Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/91925
Type: Artigo de evento
Title: Evolving Gradient A New Approach To Perform Neural Network Training
Author: Berci C.D.
Bottura C.P.
Abstract: The use of genetic algorithms in ANNs training is not a new subject, several works have already accomplished good results, however not competitive with procedural methods for problems where the gradient of the error is well defined. The present document proposes an alternative for ANNs training using GA(Genetic Algorithms) to evolve the training process itself and not to evolve directly the network parameters. This way we get quite superior results and obtain a method competitive with these, usually used to training ANNs.
Editor: 
Rights: fechado
Identifier DOI: 
Address: http://www.scopus.com/inward/record.url?eid=2-s2.0-74549132465&partnerID=40&md5=33ac4e90272dd11305f45a5fbe1afea0
Date Issue: 2009
Appears in Collections:Unicamp - Artigos e Outros Documentos

Files in This Item:
File Description SizeFormat 
2-s2.0-74549132465.pdf1.18 MBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.