A Neural Learning Rule for CCA Approximation

  • M. Shahjahan
  • K. Murase
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3316)

Abstract

A new learning rule is implemented for approximating canonical correlation analysis(CCA) with artificial neural networks. A correlation objective function is maximized in order to find identical or correlated item from several sets of data. A simple weight update rule is derived, that is computationally much more inexpensive than the standard statistical technique. We demonstrate the network capabilities on artificial and real-world data. The experimental results show that this method is a good approximator of CCA as well as correlated item identifier.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Charles, D., Fyfe, C.: Modeling multiple cause structure using rectification constraints. Networks: Computation in Neural Systems 9, 167–182 (1998)MATHCrossRefGoogle Scholar
  2. Fyfe, C., Baddeley, R.: Non-linear data structure extracting using simple habbian networks. Biological Cybernetics 72(6), 533–541 (1995)MATHCrossRefGoogle Scholar
  3. Hyverinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. John Wiley & Sons Publishing, USA (2001)CrossRefGoogle Scholar
  4. Lai, P.L., Fyfe, C.: A neural implementation of canonical correlation analysis. Neural Networks 12, 1391–1397 (1999)CrossRefGoogle Scholar
  5. Mardia, K.V., Kent, J., Bibby: Multivariate analysis. Academic Press, New York (1979)MATHGoogle Scholar
  6. Oja, E.: A simplified neuron model as a principle component analyzer. Journal of Mathematical Biology 16, 267–273 (1982)CrossRefMathSciNetGoogle Scholar
  7. Oja, E.: The nonlinear pca learning rule in independent component analysis. Neurocomputing 17, 25–45 (1997)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • M. Shahjahan
    • 1
  • K. Murase
    • 1
  1. 1.Department of Human and Artificial Intelligence SystemsUniversity of FukuiFukuiJapan

Personalised recommendations