MULTI-REGULARIZATION PARAMETERS ESTIMATION FOR GAUSSIAN MIXTURE CLASSIFIER BASED ON MDL PRINCIPLE

Xiuling Zhou, Ping Guo, C. L. Philip Chen

2011

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.

Paper Citation

in Harvard Style

Zhou X., Guo P. and Chen C. (2011). MULTI-REGULARIZATION PARAMETERS ESTIMATION FOR GAUSSIAN MIXTURE CLASSIFIER BASED ON MDL PRINCIPLE . In Proceedings of the International Conference on Neural Computation Theory and Applications - Volume 1: NCTA, (IJCCI 2011) ISBN 978-989-8425-84-3, pages 112-117. DOI: 10.5220/0003669301120117

in Bibtex Style

@conference{ncta11,
author={Xiuling Zhou and Ping Guo and C. L. Philip Chen},
title={MULTI-REGULARIZATION PARAMETERS ESTIMATION FOR GAUSSIAN MIXTURE CLASSIFIER BASED ON MDL PRINCIPLE},
booktitle={Proceedings of the International Conference on Neural Computation Theory and Applications - Volume 1: NCTA, (IJCCI 2011)},
year={2011},
pages={112-117},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0003669301120117},
isbn={978-989-8425-84-3},
}

in EndNote Style

TY - CONF
JO - Proceedings of the International Conference on Neural Computation Theory and Applications - Volume 1: NCTA, (IJCCI 2011)
TI - MULTI-REGULARIZATION PARAMETERS ESTIMATION FOR GAUSSIAN MIXTURE CLASSIFIER BASED ON MDL PRINCIPLE
SN - 978-989-8425-84-3
AU - Zhou X.
AU - Guo P.
AU - Chen C.
PY - 2011
SP - 112
EP - 117
DO - 10.5220/0003669301120117