IJPAM: Volume 104, No. 3 (2015)

MATRICIAL REPRESENTATION OF INDIVIDUALS IN
CONTINUOUS GENETIC ALGORITHMS CAN IMPROVE
THE EFFICIENCE IN RADIAL BASIS FUNCTIONS
NEURAL NETWORKS TRAINING

J.F. da Mota$^1$, P.H. Siqueira$^2$, L.V. de Souza$^3$
Paraná State University
733 Com. Norberto Marcondes Av.
Campo Mourão, 87303-100, BRAZIL
$^2$Federal Parana University
Polytechnic Center
Curitiba, 81531-970, BRAZIL
$^3$Federal Parana University
Polytechnic Center
Curitiba, 81531-970, BRAZIL


Abstract. In this paper we've compared two GA - Genetic Algorithm - based approaches for computing the weight matrix of a RBFNN - Radial Basis Function Neural Network. The first, named GA, was based in Michalewicz's Operators for Continuous Genetic Algorithms and the other, named modGA, was based in extending these operators to matricial individuals, consequently proposing new operators. The main objective was verifying if the new approach could reduce the number of iterations (generations) necessary to compute the weight matrix. Six datasets was tested and in 50% of them, that hypothesis was confirmed.

Received: July 30, 2015

AMS Subject Classification: 92B20

Key Words and Phrases: RBF neural networks, genetic algorithms, classification, RBF training

Download paper from here.



DOI: 10.12732/ijpam.v104i3.11 How to cite this paper?

Source: International Journal of Pure and Applied Mathematics
ISSN printed version: 1311-8080
ISSN on-line version: 1314-3395
Year: 2015
Volume: 104
Issue: 3
Pages: 421 - 436


Google Scholar; DOI (International DOI Foundation); WorldCAT.

CC BY This work is licensed under the Creative Commons Attribution International License (CC BY).