Abstract

Support vector machines (SVM's) construct decision functions that are linear combinations of kernel evaluations on the training set. The samples with non-vanishing coefficients are called support vectors. In this work we establish lower (asymptotical) bounds on the number of support vectors. On our way we prove several results which are of great importance for the understanding of SVM's. In particular, we describe to which "limit" SVM decision functions tend, discuss the corresponding notion of convergence and provide some results on the stability of SVM's using subdifferential calculus in the associated reproducing kernel Hilbert space.

Citations

801	A probabilistic theory of pattern recognition - Devroye, Györfi, et al. - 1996
741	Shawe-Taylor J (2000) An Introduction to Support Vector Machines and other kernel-based learning methods - Cristianini
560	Theory of reproducing kernels - Aronszajn - 1950
110	Central limit theorems for empirical measures - Dudley - 1979
79	Harmonic Analysis on Semigroups: Theory of Positive Definite and Related Functions - Berg, Christensen, et al. - 1984
61	Support vector machines and the Bayes rule in classification - Lin
29	Integral functionals, normal integrands and measurable selections - Rockafellar - 1975
2	A generalized representer theorem - olkopf, Herbrich, et al. - 2001
1	New results in subdierential calculus with applications to convex optimization - Romano - 1995