Learning sparse features for classification by mixture models

Weixiang Liu

doi:10.1016/j.patrec.2003.09.007

Return

Learning sparse features for classification by mixture models

2004-01-01

PRE

刘

刘维湘 (Weixiang Liu)

Nanning Zheng

DOI：10.1016/j.patrec.2003.09.007

Original

Original request for help

Save

Abstract

En 中文

Non-negative matrix factorization (NMF) can discover sparse features for classification via mixture models and the sparseness of features controls the learning rate of the basis function parameters. But the original NMF in which the basis vectors are unit ones in L-1 norm, does not increase the sparseness of learned features. This paper generalizes NMF to L-p-NMF where the basis vectors are unit ones in L-p norm. Experiments demonstrate how p affects the sparseness of learned features and the final classification accuracy. And the results show that L-2-NMF is superior one for practical implementation. (C) 2003 Elsevier B.V. All rights reserved.

Keywords:

non-negative matrix factorization

classification

mixture models

sparse features

Lp norm

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.