On the error sensitivity measure for pruning RBF networks

Author

Sum, John ; Leung, Chi-sing

Author_Institution

Dept. of Comput., Hong Kong Polytech. Univ., Hung Hom, China

Volume

2

fYear

2003

fDate

2-5 Nov. 2003

Firstpage

1162

Abstract

Error sensitivity measure is normally a commonly used factor for searching the optimal structure of a neural network. Starting with the derivation of a recursive equation for the update of a reduced order parametric vector based on the full order parametric vector, the error sensitivity measure for use in linear regressor and RBF network pruning is re-derived and an approximated error sensitivity measure identical to that of proposed in optimal brain damage has been obtained. Considering the training is accomplished by recursive least square method, an on-line training-pruning algorithm is proposed.

Keywords

learning (artificial intelligence); radial basis function networks; recursive estimation; RBF network pruning; error sensitivity measure; full order parametric vector; linear regressor; online training-pruning algorithm; optimal brain damage; recursive equation; recursive least square method; reduced order parametric vector; Cybernetics; Equations; Machine learning; Neural networks; Radial basis function networks; Signal processing; Training data;

fLanguage

English

Publisher

ieee

Conference_Titel

Machine Learning and Cybernetics, 2003 International Conference on

Print_ISBN

0-7803-8131-9

Type

conf

DOI

10.1109/ICMLC.2003.1259661

Filename

1259661