Speeding Up SVM in Test Phase: Application to Radar HRRP ATR

  • Bo Chen
  • Hongwei Liu
  • Zheng Bao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4232)


In this paper, a simple method is proposed to reduce the number of support vectors (SVs) in the decision function. Because in practice the embedded data just lie into a subspace of the kernel-induced space, F, we can search a set of basis vectors (BVs) to express all the SVs according to the geometrical structure, the number of which is less than that of SVs. The experimental results show that our method can reduce the run-time complexity in SVM with the preservation of machine’s generalization, especially for the data of large correlation coefficients among input samples.


Decision Function Generalization Performance Kernel Parameter Sequential Forward Selection Loss Generalization 
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.
    Vapnik, V.: The nature of statistical learning theory. Springer, Heidelberg (1995)MATHGoogle Scholar
  2. 2.
    Burges, C.: Simplified support vector decision rules. In: Proc. 13th International Conference on Machine Learning, San Mateo, CA, pp. 71–77 (1996)Google Scholar
  3. 3.
    Schoelkopf, B., Mika, S., Burges, C., Knirsch, P., Muller, K., Ratsch, G., Smola, A.: Input space versus feature space in kernel-based methods. IEEE Trans. Neural Networks 10, 1000–1017 (1999)CrossRefGoogle Scholar
  4. 4.
    Nguyen, D., Ho, T.: An efficient method for simplifying support machines. In: Proc. 22nd ICML, Bonn, Germany (2005)Google Scholar
  5. 5.
    Baudat, G., Anouar, F.: Feature vector selection and projection using kernels. Neurocomputing 5, 20–38 (2003)Google Scholar
  6. 6.
    Bengio, Y., Delalleau, O., Nicolas, L.: The curse of dimensionality for local kernel machines. Technical Report 1258, Dept. IRO, University of Montreal, Canada (2005)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Bo Chen
    • 1
  • Hongwei Liu
    • 1
  • Zheng Bao
    • 1
  1. 1.National Lab of Radar Signal ProcessingXidian UniversityXi’an, ShaanxiP.R. China

Personalised recommendations