An experimental study: on reducing RBF input dimension by ICA and PCA

Rong Bo Huang, Lap Tak Law, Yiu Ming Cheung

Research output: Chapter in book/report/conference proceedingConference proceedingpeer-review

14 Citations (Scopus)

Abstract

Experimentally investigates using independent component analysis (ICA) and principle component analysis (PCA) in the reduction of the input dimension of a radial basis function (RBF) network such that the net's complexity is reduced. The results have shown that a RBF network with ICA as an input pre-process has similar generalization ability to the one without pre-processing, but the former's performance converges much faster. In contrast, a PCA based RBF leads to a deteriorated result in both convergent speed and generalization ability.

Original languageEnglish
Title of host publicationProceedings of 2002 International Conference on Machine Learning and Cybernetics
PublisherIEEE
Pages1941-1945
Number of pages5
Volume4
DOIs
Publication statusPublished - Nov 2002
Event2002 International Conference on Machine Learning and Cybernetics - Beijing, China
Duration: 4 Nov 20025 Nov 2002

Publication series

NameInternational Conference on Machine Learning and Cybernetics, ICMLC

Conference

Conference2002 International Conference on Machine Learning and Cybernetics
Country/TerritoryChina
CityBeijing
Period4/11/025/11/02

Fingerprint

Dive into the research topics of 'An experimental study: on reducing RBF input dimension by ICA and PCA'. Together they form a unique fingerprint.

Cite this