An EM algorithm for learning sparse and overcomplete representations

Mingjun Zhong, Huanwen Tang, Hongjun Chen, Yiyuan Tang

Research output: Contribution to journalArticlepeer-review

11 Scopus citations

Abstract

An expectation-maximization (EM) algorithm for learning sparse and overcomplete representations is presented in this paper. We show that the estimation of the conditional moments of the posterior distribution can be accomplished by maximum a posteriori estimation. The approximate conditional moments enable the development of an EM algorithm for learning the overcomplete basis vectors and inferring the most probable basis coefficients.

Original languageEnglish (US)
Pages (from-to)469-476
Number of pages8
JournalNeurocomputing
Volume57
Issue number1-4
DOIs
StatePublished - Mar 2004
Externally publishedYes

Keywords

  • EM algorithm
  • Maximum a posteriori
  • Overcomplete representations

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence

Fingerprint

Dive into the research topics of 'An EM algorithm for learning sparse and overcomplete representations'. Together they form a unique fingerprint.

Cite this