Non-logarithmic information measures, α-weighted em algorithms and speedup of learning

IEEE International Symposium on Information Theory - Proceedings

Volumn , Issue , 1998, Pages 385-

  Matsuyama, Y ^a,b  

^a WASEDA UNIVERSITY   (Japan)

^b RWCP ^*  (Japan)

Author keywords

[No Author keywords available]

Indexed keywords

ACCELERATION OF CONVERGENCE; EM ALGORITHMS; GAUSSIAN MIXTURES; GENERALIZED EM ALGORITHMS; INFORMATION MEASURES;

CRAMER-RAO BOUNDS; INFORMATION THEORY;

ALGORITHMS;

EID: 84871368049     PISSN: 21578095     EISSN: None     Source Type: Conference Proceeding    
DOI: 10.1109/ISIT.1998.708990     Document Type: Conference Paper

References (3)

1. A.P. Dempster, N.M. Laird and D.B. Rubin, "Maximum likelihood from incomplete data via the EM algorithm," J. R. Stat. Sr. B, vol. 39, pp. 1-38, 1978.
2. Y. Matsuyama, "The a-EM algorithm," Lecture Notes in Computer Science 1240, pp. 483-492, Springer, 1997.
3. Y. Matsuyama, "WEM algorithms and probabilistic learning," Proc. 29th ISCIE-SSS, Ohm, pp. 1-12, 1997.