Speeding Up SVM in Test Phase: Application to Radar HRRP ATR
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.
KeywordsDecision Function Generalization Performance Kernel Parameter Sequential Forward Selection Loss Generalization
Unable to display preview. Download preview PDF.
- 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
- 4.Nguyen, D., Ho, T.: An efficient method for simplifying support machines. In: Proc. 22nd ICML, Bonn, Germany (2005)Google Scholar
- 5.Baudat, G., Anouar, F.: Feature vector selection and projection using kernels. Neurocomputing 5, 20–38 (2003)Google Scholar
- 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