TY - JOUR
T1 - On weight design of maximum weighted likelihood and an extended EM algorithm
AU - Zhang, Zhenyue
AU - CHEUNG, Yiu Ming
N1 - Funding Information:
The authors would like to sincerely thank the editor and the anonymous reviewers for their valuable comments and insightful suggestions. They would also like to thank Mr. Xing-ming Zhao for conducting Experiment 5. The work described in this paper was supported by Faculty Research Grant of Hong Kong Baptist University with the Project Code: FRG/05-06/II-42, by the Research Grant Council of Hong Kong SAR under Project HKBU 2156/04E, and in part by the Special Funds for Major State Basic Research Projects (project G19990328) and NSFC (project 60372033).
PY - 2006/10
Y1 - 2006/10
N2 - The recent Maximum Weighted Likelihood (MWL) [18], [19] has provided a general learning paradigm for density-mixture model selection and learning, in which weight design, however, is a key issue. This paper will therefore explore such a design, and through which a heuristic extended Expectation-Maximization (X-EM) algorithm is presented accordingly. Unlike the EM algorithm [1], the X-EM algorithm is able to perform model selection by fading the redundant components out from a density mixture, meanwhile estimating the model parameters appropriately. The numerical simulations demonstrate the efficacy of our algorithm.
AB - The recent Maximum Weighted Likelihood (MWL) [18], [19] has provided a general learning paradigm for density-mixture model selection and learning, in which weight design, however, is a key issue. This paper will therefore explore such a design, and through which a heuristic extended Expectation-Maximization (X-EM) algorithm is presented accordingly. Unlike the EM algorithm [1], the X-EM algorithm is able to perform model selection by fading the redundant components out from a density mixture, meanwhile estimating the model parameters appropriately. The numerical simulations demonstrate the efficacy of our algorithm.
KW - Extended expectation-maximization algorithm
KW - Maximum weighted likelihood
KW - Model selection
KW - Weight design
UR - http://www.scopus.com/inward/record.url?scp=33748376519&partnerID=8YFLogxK
U2 - 10.1109/TKDE.2006.163
DO - 10.1109/TKDE.2006.163
M3 - Journal article
AN - SCOPUS:33748376519
SN - 1041-4347
VL - 18
SP - 1429
EP - 1434
JO - IEEE Transactions on Knowledge and Data Engineering
JF - IEEE Transactions on Knowledge and Data Engineering
IS - 10
M1 - 1683776
ER -