Bayesian ying-yang system and theory as a unified statistical learning approach: (III) models and algorithms for dependence reduction, data dimension reduction, ICA and supervised learning

摘要

This paper is a sister paper of [1] published in this same proceeding, for further interpreting the Bayesian Ying-Yang (BYY) learning system and theory through its uses on developing models and algorithms for dependence reduction, independent component analysis, data dimension reduction, supervised classification and regression with three-layer net, mixtures-of-experts, and radial basis function nets. Readers are referred to [14,1] for the details on BYY lerning system and theory. In addition, the relation of BYY learning system and theory to a number of existing learning model and theories has been discussed in [14].