TRAINING RADIAL BASIS FUNCTION NETWORKS BY GENETIC ALGORITHMS

Juliano F. da Mota, Paulo H. Siqueira, Luzia V. de Souza, Adriano Vitor

2012

Abstract

One of the issues of modeling a RBFNN - Radial Basis Function Neural Network consists of determining the weights of the output layer, usually represented by a rectangular matrix. The inconvenient characteristic at this stage it’s the calculation of the pseudo-inverse of the activation values matrix. This operation may become computationally expensive and cause rounding errors when the amount of variables is large or the activation values form an ill-conditioned matrix so that the model can misclassify the patterns. In our research, Genetic Algorithms for continuous variables determines the weights of the output layer of a RBNN and we’ve made a comparsion with the traditional method of pseudo-inversion. The proposed approach generates matrices of random normally distributed weights which are individuals of the population and applies the Michalewicz’s genetic operators until some stopping criteria is reached. We’ve tested four classification patterns databases and an overall mean accuracy lies in the range 91–98%, in the best case and 58–63%, in the worse case.

Paper Citation

in Harvard Style

F. da Mota J., H. Siqueira P., V. de Souza L. and Vitor A. (2012). TRAINING RADIAL BASIS FUNCTION NETWORKS BY GENETIC ALGORITHMS . In Proceedings of the 4th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART, ISBN 978-989-8425-95-9, pages 373-379. DOI: 10.5220/0003751903730379

in Bibtex Style

@conference{icaart12,
author={Juliano F. da Mota and Paulo H. Siqueira and Luzia V. de Souza and Adriano Vitor},
title={TRAINING RADIAL BASIS FUNCTION NETWORKS BY GENETIC ALGORITHMS},
booktitle={Proceedings of the 4th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,},
year={2012},
pages={373-379},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003751903730379},
isbn={978-989-8425-95-9},
}

in EndNote Style

TY - CONF
JO - Proceedings of the 4th International Conference on Agents and Artificial Intelligence - Volume 1: ICAART,
TI - TRAINING RADIAL BASIS FUNCTION NETWORKS BY GENETIC ALGORITHMS
SN - 978-989-8425-95-9
AU - F. da Mota J.
AU - H. Siqueira P.
AU - V. de Souza L.
AU - Vitor A.
PY - 2012
SP - 373
EP - 379
DO - 10.5220/0003751903730379