DocumentCode
2498325
Title
Acceleration of the EM algorithm via proximal point iterations
Author
Chretien, Stdphane ; Hero, Alfred O., III
Author_Institution
Dept. of Electr. Eng. & Comput. Sci., Michigan Univ., Ann Arbor, MI, USA
fYear
1998
fDate
16-21 Aug 1998
Firstpage
444
Abstract
In this paper, a class of proximal point methods using the Kullback information measure is introduced. In particular, the EM algorithm is proved to belong to this class. A novel relaxation of the EM algorithm is proposed along with implementation strategies. Superlinear convergence of these methods is established
Keywords
convergence of numerical methods; information theory; iterative methods; maximum likelihood estimation; EM algorithm acceleration; Kullback information measure; expectation maximization algorithm; implementation strategies; proximal point iterations; relaxation; superlinear convergence; Acceleration; Approximation algorithms; Computer science; Convergence; Data models; Electric variables measurement; Maximum likelihood estimation; Minimization methods; Particle measurements; Random variables;
fLanguage
English
Publisher
ieee
Conference_Titel
Information Theory, 1998. Proceedings. 1998 IEEE International Symposium on
Conference_Location
Cambridge, MA
Print_ISBN
0-7803-5000-6
Type
conf
DOI
10.1109/ISIT.1998.709049
Filename
709049
Link To Document