Published March 2001
| Published
Journal Article
Open
A Constrained EM Algorithm for Independent Component Analysis
- Creators
- Welling, Max
- Weber, Markus
Abstract
We introduce a novel way of performing independent component analysis using a constrained version of the expectation-maximization (EM) algorithm. The source distributions are modeled as D one-dimensional mixtures of gaussians. The observed data are modeled as linear mixtures of the sources with additive, isotropic noise. This generative model is fit to the data using constrained EM. The simpler "soft-switching" approach is introduced, which uses only one parameter to decide on the sub- or supergaussian nature of the sources. We explain how our approach relates to independent factor analysis.
Additional Information
© 2001 Massachusetts Institute of Technology. Received March 15, 1999; accepted June 1, 2000. Posted Online March 13, 2006. We thank Pietro Perona for stimulating discussions and the referees for many suggestions that improved the text significantly. M. Welling acknowledges the Sloan Center for its ongoing financial support.Attached Files
Published - WELnc01.pdf
Files
WELnc01.pdf
Files
(266.6 kB)
| Name | Size | Download all |
|---|---|---|
|
md5:1d42272f178ebe6d0563c483a1676e38
|
266.6 kB | Preview Download |
Additional details
- Eprint ID
- 27691
- Resolver ID
- CaltechAUTHORS:20111109-084139616
- Sloan Center
- Created
-
2012-01-18Created from EPrint's datestamp field
- Updated
-
2021-11-09Created from EPrint's last_modified field