Display options
Share it on

J Mach Learn Res. 2002 Aug 01;3:99-114.

Variational Learning of Clusters of Undercomplete Nonsymmetric Independent Components.

Journal of machine learning research : JMLR

Kwokleung Chan, Te-Won Lee, Terrence J Sejnowski

Affiliations

  1. Computational Neurobiology Laboratory, The Salk Institute, 10010 North Torrey Pines Road, La Jolla, CA 92037, USA.

PMID: 21479123 PMCID: PMC3072251

Abstract

We apply a variational method to automatically determine the number of mixtures of independent components in high-dimensional datasets, in which the sources may be nonsymmetrically distributed. The data are modeled by clusters where each cluster is described as a linear mixture of independent factors. The variational Bayesian method yields an accurate density model for the observed data without overfitting problems. This allows the dimensionality of the data to be identified for each cluster. The new method was successfully applied to a difficult real-world medical dataset for diagnosing glaucoma.

References

  1. IEEE Trans Biomed Eng. 2002 Sep;49(9):963-74 - PubMed
  2. Neural Comput. 1999 Nov 15;11(8):1875-83 - PubMed
  3. Neural Comput. 2001 Mar;13(3):677-89 - PubMed
  4. Nature. 1996 Jun 13;381(6583):607-9 - PubMed
  5. Neural Comput. 2002 Apr;14(4):889-918 - PubMed
  6. Proc IEEE Inst Electr Electron Eng. 2001 Jul 1;89(7):1107-1122 - PubMed
  7. Br J Ophthalmol. 1977 Feb;61(2):107-13 - PubMed
  8. Neural Comput. 1999 May 15;11(4):803-51 - PubMed

Publication Types

Grant support