Expectation-Maximization approaches to independent component analysis

Mingjun Zhong, Huanwen Tang, Yiyuan Tang

Research output: Contribution to journalArticlepeer-review

5 Scopus citations


Expectation-Maximization (EM) algorithms for independent component analysis are presented in this paper. For super-Gaussian sources, a variational method is employed to develop an EM algorithm in closed form for learning the mixing matrix and inferring the independent components. For sub-Gaussian sources, a symmetrical form of the Pearson mixture model (Neural Comput. 11 (2) (1999) 417-441) is used as the prior, which also enables the development of an EM algorithm in fclosed form for parameter estimation.

Original languageEnglish (US)
Pages (from-to)503-512
Number of pages10
Issue number1-4
StatePublished - Oct 2004
Externally publishedYes


  • EM algorithm
  • Independent component analysis
  • Overcomplete representations
  • Variational method

ASJC Scopus subject areas

  • Computer Science Applications
  • Cognitive Neuroscience
  • Artificial Intelligence


Dive into the research topics of 'Expectation-Maximization approaches to independent component analysis'. Together they form a unique fingerprint.

Cite this