An Empirical Comparison of Training Algorithms for Radial Basis Functions

Authors:
Mamen Ortiz-Gómez;Carlos Hernández-Espinosa;Mercedes Fernández-Redondo
Affiliations:
Universidad Jaume I, Castellón, Spain 12071;Universidad Jaume I, Castellón, Spain 12071;Universidad Jaume I, Castellón, Spain 12071
Venue:
IWANN '03 Proceedings of the 7th International Work-Conference on Artificial and Natural Neural Networks: Part II: Artificial Neural Nets Problem Solving Methods
Year:
2009

Citing 7
Cited 0

Original Contribution: On the training of radial basis function classifiers

Neural Networks
An algorithm to generate radial basis function (RBF)-like nets for classification problems

Neural Networks
An efficient method to construct a radial basis function neural network classifier

Neural Networks
Fast learning in networks of locally-tuned processing units

Neural Computation
A neural-network learning theory and a polynomial time RBF algorithm

IEEE Transactions on Neural Networks
A hybrid linear/nonlinear training algorithm for feedforward neural networks

IEEE Transactions on Neural Networks
Reformulated radial basis neural networks trained by gradient descent

IEEE Transactions on Neural Networks

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper we present a review and comparison of five different algorithms for training a RBF network. The algorithms are compared using nine databases. Our results show that the simplest algorithm, k-means clustering, may be the best alternative. The results of RBF are also compared with the results of Multilayer Feedforward with Backpropagation, the performance of a RBF network trained with k-means clustering is slightly better and the computational cost considerably lower. So we think that RBF may be a better alternative.