论文信息 - Bounds on Error Expectation for Support Vector Machines

Bounds on Error Expectation for Support Vector Machines

We introduce the concept of span of support vectors (SV) and show that the generalization ability of support vector machines (SVM) depends on this new geometrical concept. We prove that the value of the span is always smaller (and can be much smaller) than the diameter of the smallest sphere containing the support vectors, used in previous bounds (Vapnik, 1998). We also demonstate experimentally that the prediction of the test error given by the span is very accurate and has direct application in model selection (choice of the optimal parameters of the SVM).

[1] Bernhard E. Boser,et al. A training algorithm for optimal margin classifiers , 1992, COLT '92.

[2] John Shawe-Taylor,et al. Structural Risk Minimization Over Data-Dependent Hierarchies , 1998, IEEE Trans. Inf. Theory.

[3] Vladimir Vapnik,et al. Statistical learning theory , 1998 .

[4] John Shawe-Taylor,et al. Generalization Performance of Support Vector Machines and Other Pattern Classifiers , 1999 .

[5] B. Schölkopf,et al. Advances in kernel methods: support vector learning , 1999 .

[6] R. C. Williamson,et al. Generalization Bounds via Eigenvalues of the Gram matrix , 1999 .

[7] Vladimir N. Vapnik,et al. The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[8] Gunnar Rätsch,et al. An introduction to kernel-based learning algorithms , 2001, IEEE Trans. Neural Networks.

[9] Peter Williams,et al. The generalization error of the symmetric and scaled support vector machines , 2001, IEEE Trans. Neural Networks.