The latent maximum entropy principle

被引：8

作者：

Wang, SJ ^{[1
]}

Rosenfeld, R ^{[1
]}

Zhao, YX ^{[1
]}

Schuurmans, D ^{[1
]}

机构：

[1] Carnegie Mellon Univ, Sch Comp Sci, Pittsburgh, PA 15213 USA

来源：

ISIT: 2002 IEEE INTERNATIONAL SYMPOSIUM ON INFORMATION THEORY, PROCEEDINGS | 2002年

关键词：

D O I：

10.1109/ISIT.2002.1023403

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

We present an extension of Jaynes' maximum entropy principle to handle latent variables. We use an EM algorithm that incorporates nested iterative scaling to approximately calculate maximum entropy solutions for this principle, and give a proof of its convergence.

引用

页码：131 / 131

页数：1

共 4 条

[1] GENERALIZED ITERATIVE SCALING FOR LOG-LINEAR MODELS [J].

DARROCH, JN ;

RATCLIFF, D .

ANNALS OF MATHEMATICAL STATISTICS, 1972, 43 (05) :1470-&

[2] Inducing features of random fields [J].

DellaPietra, S ;

DellaPietra, V ;

Lafferty, J .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1997, 19 (04) :380-393

[3] MAXIMUM LIKELIHOOD FROM INCOMPLETE DATA VIA EM ALGORITHM [J].

DEMPSTER, AP ;

LAIRD, NM ;

RUBIN, DB .

JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-METHODOLOGICAL, 1977, 39 (01) :1-38

[4]

Jaynes E. T., 1983, Papers on Probability, Statistics and Statistical Physics

← 1 →