View Publication

The latent maximum entropy principle

Shaojun Wang, Dept of Computing Science
Ronald Rosenfeld, School of Computer Science, Carnegie Mellon University
Yunxin Zhao, Department of Computer Engineering and Computer Science, University of Missouri at Columbia
Dale Schuurmans, AICML

We present an extension to Jaynes' maximum entropy principle that handles latent variables. The principle of latent maximum entropy we propose is di#erent from both Jaynes' maximum entropy principle and maximum likelihood estimation, but often yields better estimates in the presence of hidden variables and limited training data. We first show that solving for a latent maximum entropy model poses a hard nonlinear constrained optimization problem in general. However, we then show that

Citation

S. Wang, R. Rosenfeld, Y. Zhao, D. Schuurmans. "The latent maximum entropy principle". International Symposium on Information Theory (ISIT), June 2002.

Keywords:
Category:	In Conference

BibTeX

@incollection{Wang+al:ISIT02,
  author = {Shaojun Wang and Ronald Rosenfeld and Yunxin Zhao and Dale
    Schuurmans},
  title = {The latent maximum entropy principle},
  booktitle = {International Symposium on Information Theory (ISIT)},
  year = 2002,
}

Last Updated: June 01, 2007
Submitted by Staurt H. Johnson

Not Logged In

PapersDB

The latent maximum entropy principle

Citation

BibTeX