Title :
On Weight Design of Maximum Weighted Likelihood and an Extended EM Algorithm
Author :
Zhang, Zhenyue ; Cheung, Yiu-Ming
Author_Institution :
Dept. of Math., Zhejiang Univ.
Abstract :
The recent maximum weighted likelihood (MWL) 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, 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
Keywords :
expectation-maximisation algorithm; learning (artificial intelligence); density-mixture model selection; expectation-maximization algorithm; extended EM algorithm; maximum weighted likelihood; Algorithm design and analysis; Clustering algorithms; Convergence; Expectation-maximization algorithms; Fading; Hidden Markov models; Iterative algorithms; Maximum likelihood estimation; Numerical simulation; Parameter estimation; Maximum weighted likelihood; extended expectation-maximization algorithm; model selection.; weight design;
Journal_Title :
Knowledge and Data Engineering, IEEE Transactions on
DOI :
10.1109/TKDE.2006.163
