Publication Details

Neural Network Training using Genetic Algorithm

GAŽAR Roman. Neural Network Training using Genetic Algorithm. In: Mendel 2004 10th Internacional Conference on Soft Computing. Brno: Faculty of Mechanical Engineering BUT, 2004, pp. 161-166. ISBN 80-214-2676-4.
Czech title
Učení neuronových sítí s využitím genetických algoritmů
Type
conference paper
Language
english
Authors
Gažar Roman, Mgr. (DCSY FIT BUT)
Keywords

neural network, genetic algorithm, training set, weight selection

Abstract

This paper deals with the aggregation of  two well known computing paradigm  - neural networks (NN) and genetic algorithms (GA). GA in NN area can be used for searching weight values, topology design, NN parameter settings and for selection and ordering of input and output vectors for training and testing set. We focused only on the weight  searching by GA. The structure of NN is fixed and it is not changed throw all experiments. The feed-forward NN is used with one hidden layer and standard sigmoidal transfer function. We used standard genetic algorithm with  external setting of its parameters

Published
2004
Pages
161-166
Proceedings
Mendel 2004 10th Internacional Conference on Soft Computing
Conference
Tenth International Mendel Conference on Soft Computing, FME, VUT BRNO, CZ
ISBN
80-214-2676-4
Publisher
Faculty of Mechanical Engineering BUT
Place
Brno, CZ
BibTeX 
@INPROCEEDINGS{FITPUB7522,
   author = "Roman Ga\v{z}ar",
   title = "Neural Network Training using Genetic Algorithm",
   pages = "161--166",
   booktitle = "Mendel 2004 10th Internacional Conference on Soft Computing",
   year = 2004,
   location = "Brno, CZ",
   publisher = "Faculty of Mechanical Engineering BUT",
   ISBN = "80-214-2676-4",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/7522"
}
Back to top