Givens rotation based fast backward elimination algorithm for RBF neural network pruning

Hong, X. ORCID: https://orcid.org/0000-0002-6832-2298 and Billings, S.A. (1997) Givens rotation based fast backward elimination algorithm for RBF neural network pruning. IEE Proceedings-Control Theory and Applications, 144 (5). pp. 381-384. ISSN 1350-2379 doi: 10.1049/ip-cta:19971436

Abstract/Summary

A fast backward elimination algorithm is introduced based on a QR decomposition and Givens transformations to prune radial-basis-function networks. Nodes are sequentially removed using an increment of error variance criterion. The procedure is terminated by using a prediction risk criterion so as to obtain a model structure with good generalisation properties. The algorithm can be used to postprocess radial basis centres selected using a k-means routine and, in this mode, it provides a hybrid supervised centre selection approach.