Generalized Derivative Based Kernelized Learning Vector Quantization

  • Frank-Michael Schleif
  • Thomas Villmann
  • Barbara Hammer
  • Petra Schneider
  • Michael Biehl
Part of the Lecture Notes in Computer Science book series (LNCS, volume 6283)

Abstract

We derive a novel derivative based version of kernelized Generalized Learning Vector Quantization (KGLVQ) as an effective, easy to interpret, prototype based and kernelized classifier. It is called D-KGLVQ and we provide generalization error bounds, experimental results on real world data, showing that D-KGLVQ is competitive with KGLVQ and the SVM on UCI data and additionally show that automatic parameter adaptation for the used kernels simplifies the learning.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Bartlett, P., Mendelson, S.: Rademacher and Gaussian complexities: risk bounds and structural results. Journal of Machine Learning Research 3, 463–482 (2003)MATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    Biehl, M., Hammer, B., Schleif, F.M., Schneider, P., Villmann, T.: Stationarity of matrix relevance learning vector quantization. Machine Learning Reports 3, 1–17 (2009) ISSN:1865-3960, http://www.uni-leipzig.de/~compint/mlr/mlr_01_2009.pdf Google Scholar
  3. 3.
    Bishop, C.: Pattern Recognition and Machine Learning. Springer, NY (2006)MATHGoogle Scholar
  4. 4.
    Blake, C., Merz, C.: UCI repository of machine learning databases. Department of Information and Computer Science. University of California, Irvine (1998), http://www.ics.uci.edu/~mlearn/MLRepository.html Google Scholar
  5. 5.
    Crammer, K., Gilad-Bachrach, R., Navot, A., Tishby, A.: Margin analysis of the LVQ algorithm. In: Proc. NIPS 2002, pp. 462–469 (2002)Google Scholar
  6. 6.
    Hammer, B., Villmann, T.: Generalized relevance learning vector quantization. Neural Networks 15(8-9), 1059–1068 (2002)CrossRefGoogle Scholar
  7. 7.
    Hammer, B., Villmann, T.: Mathematical aspects of neural networks. In: Verleysen, M. (ed.) Proc. of European Symposium on Artificial Neural Networks (ESANN 2003), d-side, Brussels, Belgium, pp. 59–72 (2003)Google Scholar
  8. 8.
    Kohonen, T.: Self-Organizing Maps. Springer Series in Information Sciences, vol. 30. Springer, Heidelberg (1995) (Second Extended Edition 1997) Google Scholar
  9. 9.
    Qin, A.K., Suganthan, P.N.: A novel kernel prototype-based learning algorithm. In: Proc. of ICPR’04, pp. 2621–2624 (2004)Google Scholar
  10. 10.
    Sato, A.S., Yamada, K.: Generalized learning vector quantization. In: Tesauro, G., Touretzky, D., Leen, T. (eds.) Advances in Neural Information Processing Systems, vol. 7, pp. 423–429. MIT Press, Cambridge (1995)Google Scholar
  11. 11.
    Schleif, F.M., Villmann, T., Kostrzewa, M., Hammer, B., Gammerman, A.: Cancer informatics by prototype-networks in mass spectrometry. Artificial Intelligence in Medicine 45, 215–228 (2009)CrossRefGoogle Scholar
  12. 12.
    Schneider, P., Biehl, M., Hammer, B.: Adaptive relevance matrices in learning vector quantization. Neural Computation (to appear)Google Scholar
  13. 13.
    Seo, S., Obermayer, K.: Soft learning vector quantization. Neural Computation 15, 1589–1604 (2003)MATHCrossRefGoogle Scholar
  14. 14.
    Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis and Discovery. Cambridge University Press, Cambridge (2004)Google Scholar
  15. 15.
    Simmuteit, S., Schleif, F.M., Villmann, T., Elssner, T.: Tanimoto metric in tree-som for improved representation of mass spectrometry data with an underlying taxonomic structure. In: Proc. of ICMLA 2009, pp. 563–567. IEEE Press, Los Alamitos (2009)Google Scholar
  16. 16.
    Villmann, T., Schleif, F.M.: Functional vector quantization by neural maps. In: Proceedings of Whispers 2009, p. CD (2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Frank-Michael Schleif
    • 1
  • Thomas Villmann
    • 2
  • Barbara Hammer
    • 1
  • Petra Schneider
    • 3
  • Michael Biehl
    • 3
  1. 1.Dept. of TechnUniv. of BielefeldBielefeldGermany
  2. 2.Faculty of Math./Natural and CSUniv. of Appl. Sc. MittweidaMittweidaGermany
  3. 3.Johann Bernoulli Inst. for Math. and CSUniv. of GroningenGroningenThe Netherlands

Personalised recommendations