Advertisement

On the Basis Updating Rule of Adaptive-Subspace Self-Organizing Map (ASSOM)

  • Huicheng Zheng
  • Christophe Laurent
  • Grégoire Lefebvre
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4131)

Abstract

This paper gives other views on the basis updating rule of the ASSOM proposed by Kohonen. We first show that the traditional basis vector rotation rule can be expressed as a correction to the basis vector which is a scaling of component vectors in the episode. With the latter form, some intermediate computations can be reused, leading to a computational load only linear to the input dimension and the subspace dimension, whereas a naive implementation of the traditional rotation rule has a computational load quadratic to the input dimension. We then proceed to propose a batch-mode updating of the basis vectors. We show that the correction made to each basis vector is a linear combination of component vectors in the input episode. Computations can be further saved. Experiments show that the proposed methods preserve the ability to generate topologically ordered invariant-feature filters and that the learning procedure is largely boosted.

Keywords

Basis Vector Component Vector Computational Load Learning Procedure Input Dimension 
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.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Kohonen, T.: The Adaptive-Subspace SOM (ASSOM) and its use for the implementation of invariant feature detection. In: Fogelman-Soulié, F., Gallinari, P. (eds.) Proc. ICANN 1995, Int. Conf. on Artificial Neural Networks, vol. 1, pp. 3–10. Paris (1995)Google Scholar
  2. 2.
    Kohonen, T.: Self-Organizing Maps, 3rd edn. Springer, Heidelberg (2001)MATHGoogle Scholar
  3. 3.
    Oja, E.: Principal components, minor components, and linear neural networks. Neural Networks 5, 927–935 (1992)CrossRefGoogle Scholar
  4. 4.
    Kohonen, T., Kaski, S., Lappalainen, H.: Self-organized formation of various invariant-feature filters in the Adaptive-Subspace SOM. Neural Computation 9(6), 1321–1344 (1997)CrossRefGoogle Scholar
  5. 5.
    Hase, H., Matsuyama, H., Tokutaka, H., Kishida, S.: Speech signal processing using Adaptive Subspace SOM (ASSOM). Technical Report NC95-140, The Inst. of Electronics, Information and Communication Engineers, Tottori University, Koyama, Japan (1996)Google Scholar
  6. 6.
    del Solar, J.R.: Texsom: texture segmentation using Self-Organizing Maps. Neurocomputing 21(1–3), 7–18 (1998)CrossRefGoogle Scholar
  7. 7.
    De Ridder, D., Lemmers, O., Duin, R.P., Kittler, J.: The Adaptive Subspace Map for image description and image database retrieval. In: Amin, A., Pudil, P., Ferri, F., Iñesta, J.M. (eds.) SPR 2000 and SSPR 2000. LNCS, vol. 1876, pp. 94–103. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  8. 8.
    Zhang, B., Fu, M., Yan, H., Jabri, M.: Handwritten digit recognition by Adaptive-Subspace Self-Organizing Map (ASSOM). IEEE Transactions on Neural Networks 10(4), 939–945 (1999)CrossRefGoogle Scholar
  9. 9.
    López-Rubio, E., Muñoz Pérez, J., Gómez-Ruiz, J.: A Principal Components Analysis Self-Organizing Map. Neural Networks 17, 261–270 (2004)MATHCrossRefGoogle Scholar
  10. 10.
    McGlinchey, S., Fyfe, C.: Fast formation of invariant feature maps. In: European Signal Processing Conference (EUSIPCO 1998), Island of Rhodes, Greece (1998)Google Scholar
  11. 11.
    Oja, E.: Neural networks, principal components and subspaces. International Journal of Neural Systems 1, 61–68 (1989)CrossRefMathSciNetGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Huicheng Zheng
    • 1
  • Christophe Laurent
    • 1
  • Grégoire Lefebvre
    • 1
  1. 1.France Telecom R&D – DIH/HDMCesson SévignéFrance

Personalised recommendations