Multispace KL for pattern representation and classification

This work introduces the Multispace KL (MKL) as a new approach to unsupervised dimensionality reduction for pattern representation and classification. The training set is automatically partitioned into disjoint subsets, according to an optimality criterion; each subset then determines a different KL subspace which is specialized in representing a particular group of patterns. The extension of the classical KL operators and the definition of ad hoc distances allow MKL to be effectively used where KL is commonly employed. The limits of the standard KL transform are pointed out, in particular, MKL is shown to outperform KL when the data distribution is far from a multidimensional Gaussian and to better cope with large sets of patterns, which could cause a severe performance drop in KL.