On the Kernel Widths in Radial-Basis Function Networks

Authors:
Nabil Benoudjit;Michel Verleysen
Affiliations:
Université Catholique de Louvain, Microelectronics Laboratoy, Place du Levant 3, B-1348 Louvain-la-Neuve, Belgium. e-mail: benoudjit@dice.ucl.ac.be;Université Catholique de Louvain, Microelectronics Laboratoy, Place du Levant 3, B-1348 Louvain-la-Neuve, Belgium e-mail: verleysen@dice.ucl.ac.be
Venue:
Neural Processing Letters
Year:
2003

Citing 12
Cited 15

Algorithms for better representation and faster learning in radial basis function networks

Advances in neural information processing systems 2
Universal approximation using radial-basis-function networks

Neural Computation
Vector quantization and signal compression

Vector quantization and signal compression
Original Contribution: On the training of radial basis function classifiers

Neural Networks
Approximation and radial-basis-function networks

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

Neural Networks
Radial basis function networks 2: new advances in design

Radial basis function networks 2: new advances in design
Neural Networks: A Comprehensive Foundation

Neural Networks: A Comprehensive Foundation
Neural Networks for Pattern Recognition

Neural Networks for Pattern Recognition
Pattern Recognition and Neural Networks

Pattern Recognition and Neural Networks
A new EM-based training algorithm for RBF networks

Neural Networks
On convergence properties of the em algorithm for gaussian mixtures

Neural Computation

High-dimensional delay selection for regression models with mutual information and distance-to-diagonal criteria

Neurocomputing
Resampling methods for parameter-free and robust feature selection with mutual information

Neurocomputing
Output value-based initialization for radial basis function neural networks

Neural Processing Letters
Improvement of learning algorithms for RBF neural networks in a helicopter sound identification system

Neurocomputing
Generalized multiscale radial basis function networks

Neural Networks
Simultaneous input variable and basis function selection for RBF networks

Neurocomputing
Rule extraction from differential evolution trained radial basis function network using genetic algorithms

CASE'09 Proceedings of the fifth annual IEEE international conference on Automation science and engineering
Fast bootstrap methodology for regression model selection

Neurocomputing
Input selection for radial basis function networks by constrained optimization

ICANN'07 Proceedings of the 17th international conference on Artificial neural networks
Differential evolution trained radial basis function network: application to bankruptcy prediction in banks

International Journal of Bio-Inspired Computation
Kernel Width Optimization for Faulty RBF Neural Networks with Multi-node Open Fault

Neural Processing Letters
SpectralCAT: Categorical spectral clustering of numerical and nominal data

Pattern Recognition
Simple estimate of the width in Gaussian kernel with adaptive scaling technique

Applied Soft Computing
A novel training algorithm for RBF neural network using a hybrid fuzzy clustering approach

Fuzzy Sets and Systems
Feature selection with missing data using mutual information estimators

Neurocomputing

Quantified Score

Hi-index	0.00

Visualization

Abstract

RBFN (Radial-Basis Function Networks) represent an attractive alternative to other neural network models. Their learning is usually split into an unsupervised part, where center and widths of the basis functions are set, and a linear supervised part for weight computation. Although available literature on RBFN learning widely covers how basis function centers and weights must be set, little effort has been devoted to the learning of basis function widths. This paper addresses this topic: it shows the importance of a proper choice of basis function widths, and how inadequate values can dramatically influence the approximation performances of the RBFN. It also suggests a one-dimensional searching procedure as a compromise between an exhaustive search on all basis function widths, and a non-optimal a priori choice.