Modified Kohonen’s Learning Laws for RBF Network

  • Tommi Ojala
  • Petri Vuorimaa


A hybrid training method for the radial basis function (RBF) network is presented. The method applies the Kohonen’s self-organizing map (SOM) and a modified learning vector quantization (LVQ) algorithms. Learning algorithms are derived for two alternative RBF network structures exploiting local Gaussian basis functions. The potential of the proposed methods is demonstrated by a function approximation example.


Radial Basis Function Radial Basis Function Network Output Weight Learning Vector Quantization Radial Basis Function Model 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    S. Chen, C.F. N. Cowan and P.M. Cowan: IEEE Trans Neural Networks 2, 302 (1991)CrossRefGoogle Scholar
  2. 2.
    D. Fox, V. Heinze, K. Möller, S. Thrun, and G. Veenker, Artificial Neural Networks, T. Kohonen, K. Mäkisara, O. Simula, and J. Kangas (eds.) 201, 1991.Google Scholar
  3. 3.
    J.-S. Jang: IEEE Trans. Neural Networks, 23, 665 (1993)Google Scholar
  4. 4.
    T. Kohonen: Self-organization and Associative Memory, 2nd ed., Berlin: Springer Verlag 1988.MATHGoogle Scholar
  5. 5.
    T. Kohonen: Proc. IEEE, 1464 (1990)Google Scholar
  6. 6.
    T. Kohonen, Proc. Int. Joint Conf. Neural Networks, TIJCNN-90 I, 545 (1990)Google Scholar
  7. 7.
    S. Lee and R. M. Kil: Neural Networks 4, 207 (1991)CrossRefGoogle Scholar
  8. 8.
    J. Moody and C. J. Darken: Neural Computation 1, 281 (1989)CrossRefGoogle Scholar
  9. 9.
    H. Ritter, T. Martinez, and K. Schulten: IEEE Trans. Neural Networks 1, 131 (1990)CrossRefGoogle Scholar
  10. 10.
    H. Ritter, Artificial Neural Networks, T. Kohonen, K. Mäkisara, O. Simula and J. Kangas (eds.) 379, 1991.Google Scholar
  11. 11.
    P. Vuorimaa: Fuzzy Sets and Systems 66, 223 (1994)CrossRefGoogle Scholar
  12. 12.
    L.-X. Wang and J. M. Mendel: IEEE Trans. Neural Networks 3, 807 (1992)CrossRefGoogle Scholar
  13. 13.
    IXXX studies of the self-organizing map (SOM) and learning vector quantization (LVQ). Helsinki University of Technology, Laboratory of Computer and Information Science, 1995. Available at the internet address via anonymous FTP.Google Scholar

Copyright information

© Springer-Verlag/Wien 1995

Authors and Affiliations

  • Tommi Ojala
    • 1
  • Petri Vuorimaa
    • 1
  1. 1.Signal Processing LaboratoryTampere University of TechnologyTampereFinland

Personalised recommendations