Advertisement

Fast Training of Linear Programming Support Vector Machines Using Decomposition Techniques

  • Yusuke Torii
  • Shigeo Abe
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4087)

Abstract

Decomposition techniques are used to speed up training support vector machines but for linear programming support vector machines (LP-SVMs) direct implementation of decomposition techniques leads to infinite loops. To solve this problem and to further speed up training, in this paper, we propose an improved decomposition techniques for training LP-SVMs. If an infinite loop is detected, we include in the next working set all the data in the working sets that form the infinite loop. To further accelerate training, we improve a working set selection strategy: at each iteration step, we check the number of violations of complementarity conditions and constraints. If the number of violations increases, we conclude that the important data are removed from the working set and restore the data into the working set. The computer experiments demonstrate that training by the proposed decomposition technique with improved working set selection is drastically faster than that without using the decomposition technique. Furthermore, it is always faster than that without improving the working set selection for all the cases tested.

References

  1. 1.
    Osuna, E., Freund, R., Girosi, F.: An improved training algorithm for support vector machines. In: Proc. NNSP 1997, pp. 276–285 (1997)Google Scholar
  2. 2.
    Vapnik, V.: The Nature of Statistical Learning Theory. Springer, Heidelberg (1995)zbMATHGoogle Scholar
  3. 3.
    Vapnik, V.: Statistical Learning Theory. John Wiley & Sons, Chichester (1998)zbMATHGoogle Scholar
  4. 4.
    Lin, C.-J.: On the convergence of the decomposition method for support vector machines. IEEE Trans. Neural Networks 12(6), 1288–1298 (2001)CrossRefGoogle Scholar
  5. 5.
    Keerthi, S.S., Gilbert, E.G.: Convergence of a generalized SMO algorithm for SVM classifier design. Machine Learning 46, 351–360 (2002)zbMATHCrossRefGoogle Scholar
  6. 6.
    Bennett, K.P.: Combining support vector and mathematical programming methods for classification. In: Schölkopf, B., et al. (eds.) Advances in Kernel Methods: Support Vector Learning, pp. 307–326. MIT Press, Cambridge (1999)Google Scholar
  7. 7.
    Abe, S.: Support Vector Machines for Pattern Classification. Springer, Heidelberg (2005)Google Scholar
  8. 8.
    Schölkopf, B., Smola, A.J.: Learning with Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. MIT Press, Cambridge (2002)Google Scholar
  9. 9.
  10. 10.
    Vanderbei, R.J.: Linear Programming, 2nd edn. Kluwer Academic Publishers, Dordrecht (2001)zbMATHGoogle Scholar
  11. 11.
    Chavátal, V.: Linear Programming. W.H. Freeman and Company, New York (1983)Google Scholar
  12. 12.
    Takenaga, H., et al.: Input layer optimization of neural networks by sensitivity analysis and its application to recognition of numerals. Electrical Engineering in Japan 111(4), 130–138 (1991)CrossRefGoogle Scholar
  13. 13.
    Hashizume, A., Motoike, J., Yabe, R.: Fully automated blood cell differential system and its application. In: Proc. IUPAC 3rd International Congress on Automation and New Technology in the Clinical Laboratory, pp. 297–302 (1988)Google Scholar
  14. 14.
    Weiss, S.M., Kapouleas, I.: An empirical comparison of pattern recognition, neural nets, and machine learning classification methods. In: Proc. IJCAI, pp. 781–787 (1989)Google Scholar
  15. 15.
    Lan, M.-S., Takenaga, H., Abe, S.: Character recognition using fuzzy rules extracted from data. In: Proc. 3rd IEEE International Conference on Fuzzy Systems, vol. 1, pp. 415–420 (1994)Google Scholar
  16. 16.
    Abe, S.: Pattern Classification: Neuro-Fuzzy Methods and Their Comparison. Springer, Heidelberg (2001)zbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Yusuke Torii
    • 1
  • Shigeo Abe
    • 1
  1. 1.Kobe UniversityKobeJapan

Personalised recommendations