Polynomial classifiers and support vector machines

  • Ingo Graf
  • Ulrich Kreßel
  • Jürgen Franke
Part III: Learning: Theory and Algorithms
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1327)


Polynomial support vector machines have shown a competitive performance for the problem of handwritten digit recognition. However, there is a large gap in performance vs. computing resources between the linear and the quadratic approach. By computing the complete quadratic classifier out of the quadratic support vector machine, a pivot point is found to trade between performance and effort. Different selection strategies are presented to reduce the complete quadratic classifier, which lower the required computing and memory resources by a factor of more than ten without affecting the generalization performance.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    V. Vapnik: The Nature of Statistical Learning Theory; Springer-Verlag, New York, 1995.Google Scholar
  2. 2.
    J. Schürmann: Pattern Classification: A Unified View of Statistical and Neural Approaches; John Wiley & Sons, New York, 1996.Google Scholar
  3. 3.
    R. Wilkinson et al.: The First Census Optical Character Recognition Systems Conference; National Institute of Standards and Technology (NIST), Gaithersburg, 1992.Google Scholar
  4. 4.
    U. Kreßel: The Impact of' the Learning-Set Size in Handwritten-Digit Recognition; p. 1685–1689 in T. Kohonen et al. (eds.): Artificial Neural Networks; Proceedings of the International Conference on Artificial Neural Networks (ICANN91), North-Holland, Amsterdam, 1991.Google Scholar
  5. 5.
    C. Cortes and V. Vapnik: Support-Vector Networks; Machine Learning, Nr. 20, p. 273–297, 1995.Google Scholar
  6. 6.
    R. Courant and D. Hilbert: Methods of Mathematical Physics; J. Wiley, New York, 1953.Google Scholar
  7. 7.
    I. Graf.Adaption von Klassifikatoren mit Support-Vektoren; Master Thesis, Universität Ulm, Abteilung Informationstechnik, Ulm, 1997.Google Scholar
  8. 8.
    C. Burges: Simplified Support Vector Decision Rules; Proceedings of the 13th International Conference on Machine Learning, p. 71–77, Bari, 1996.Google Scholar
  9. 9.
    U. Kreßel and J. Schürmann: Pattern Classification Techniques Based on Function Approximation; chapter 2 in [11].Google Scholar
  10. 10.
    J. Franke: Isolated Handprinted Digit Recognition; chapter 4 in [11].Google Scholar
  11. 11.
    P. Wang and H. Bunke (eds.): Handbook on Optical Character Recognition and Document Image Analysis; World Scientific Publishing Company, 1997.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • Ingo Graf
    • 1
  • Ulrich Kreßel
    • 1
  • Jürgen Franke
    • 1
  1. 1.Daimler-Benz AGResearch and TechnologyUlmGermany

Personalised recommendations