Multi-regularization parameters estimation for Gaussian mixture classifier based on MDL principle

Status	已發表Published
	Multi-regularization parameters estimation for Gaussian mixture classifier based on MDL principle
	Zhou X.1; Guo P.1; Philip Chen C.L.2
	2011-12-01
Source Publication	NCTA 2011 - Proceedings of the International Conference on Neural Computation Theory and Applications
Pages	112-117
Abstract	Regularization is a solution to solve the problem of unstable estimation of covariance matrix with a small sample set in Gaussian classifier. And multi-regularization parameters estimation is more difficult than single parameter estimation. In this paper, KLIM-L covariance matrix estimation is derived theoretically based on MDL (minimum description length) principle for the small sample problem with high dimension. KLIM-L is a generalization of KLIM (Kullback-Leibler information measure) which considers the local difference in each dimension. Under the framework of MDL principle, multi-regularization parameters are selected by the criterion of minimization the KL divergence and estimated simply and directly by point estimation which is approximated by two-order Taylor expansion. It costs less computation time to estimate the multi-regularization parameters in KLIM-L than in RDA (regularized discriminant analysis) and in LOOC (leave-one-out covariance matrix estimate) where cross validation technique is adopted. And higher classification accuracy is achieved by the proposed KLIM-L estimator in experiment.
Keyword	Covariance matrix estimation Gaussian classifier Minimum description length Multi-regularization parameters selection
URL	View the original
Language	英語English
Fulltext Access	View Full-Text via Scopus
Document Type	Conference paper
Collection	University of Macau
Affiliation	1.Beijing Normal University 2.Universidade de Macau 3.Beijing City University
Recommended Citation GB/T 7714	Zhou X.,Guo P.,Philip Chen C.L.. Multi-regularization parameters estimation for Gaussian mixture classifier based on MDL principle[C], 2011, 112-117.
APA	Zhou X.., Guo P.., & Philip Chen C.L. (2011). Multi-regularization parameters estimation for Gaussian mixture classifier based on MDL principle. NCTA 2011 - Proceedings of the International Conference on Neural Computation Theory and Applications, 112-117.

Files in This Item:
There are no files associated with this item.

If you have any objections to this item, please fill out the form below and the administrator will contact you as soon as possible.
Content:
Email：	*
Affiliation No.
Verification Code:	Refresh

Any comments and suggestions are welcomed.
Title:	*
Content:
Email：	*
Verification Code:	Refresh