Abstract

This paper studied PCA mixture model in high dimensional space. A novel EM learning approach by using perturbation was proposed for the PCA mixture model. Experiments showed the novel perturbation EM algorithm is more effective in learning PCA mixture model than an existing constrained EM algorithm.

Keywords

Mixture Model Gaussian Mixture Model Principal Direction High Dimensional Space Neural Computation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Titterington, D.M., Smith, A.F.M., Makov, U.E.: Statistical analysis of finite mixture distribution. Wiley, New York (1985)Google Scholar
  2. 2.
    McLachlan, G., Peel, D.: Fixed mixture models. Wiley, New York (2000)CrossRefGoogle Scholar
  3. 3.
    Jacobs, R., Jordan, M., Nowlan, S., Hinton, G.: Adaptive mixtures of local experts. Neural Computation 3(1), 79–87 (1991)CrossRefGoogle Scholar
  4. 4.
    Jordan, M., Jacobs, R.: Hierarchical mixtures of experts and the EM algorithm. Neural Computation 6(5), 181–214 (1994)CrossRefGoogle Scholar
  5. 5.
    Hinton, G., Dayan, P., Revow, M.: Modeling the manifolds of images of handwritten digits. IEEE Transactions on Neural Network 8(1), 65–74 (1997)CrossRefGoogle Scholar
  6. 6.
    Tipping, M.E., Bishop, C.M.: Mixtures of probabilistic principal component analysers. Neural Computation 11(2), 443–482 (1999)CrossRefGoogle Scholar
  7. 7.
    Verbeek, J.J., Vlassis, N., Kröse, B.: Coordinating mixtures of probabilistic principal component analyzers. Technical report, Computer Science Institute, University of Amsterdam, The Netherlands (Febraury 2002) IAS-UVA-02-01Google Scholar
  8. 8.
    Kim, H.-C., kim, D., Bang, S.Y.: An efficient model order selection for PCA mixture model. Pattern Recognition Letters 24(9-10), 1385–1393 (2003)MATHCrossRefGoogle Scholar
  9. 9.
    Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. Journal of the Royal Statistical Society Series B 39(1), 1–38 (1977)MATHMathSciNetGoogle Scholar
  10. 10.
    Little, R.J.A., Rubin, D.B.: Statistical analysis with missing data. Wiley, New York (1987)MATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Zhong Jin
    • 1
    • 3
  • Franck Davoine
    • 2
  • Zhen Lou
    • 3
  1. 1.Centre de Visio per ComputadorUniversitat Autonoma de BarcelonaBarcelonaSpain
  2. 2.Heudiasyc – CNRS Mixed Research UnitCompiegne University of TechnologyCompiegne cedexFrance
  3. 3.Department of Computer ScienceNanjing University of Science and TechnologyNanjingPeople’s Republic of China

Personalised recommendations