论文信息 - Shrinking the Tube: A New Support Vector Regression Algorithm

Shrinking the Tube: A New Support Vector Regression Algorithm

A new algorithm for Support Vector regression is described. For a priori chosen ν, it automatically adjusts a flexible tube of minimal radius to the data such that at most a fraction ν of the data points lie outside. Moreover, it is shown how to use parametric tube shapes with non-constant radius. The algorithm is analysed theoretically and experimentally.

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

[2] Bernhard Schölkopf,et al. Extracting Support Data for a Given Task , 1995, KDD.

[3] Bernhard Schölkopf,et al. Prior Knowledge in Support Vector Kernels , 1997, NIPS.

[4] Bernhard Schölkopf,et al. Comparing support vector machines with Gaussian kernels to radial basis function classifiers , 1997, IEEE Trans. Signal Process..

[5] R. C. Williamson,et al. Support vector regression with automatic accuracy control. , 1998 .

[6] Federico Girosi,et al. Reducing the run-time complexity of Support Vector Machines , 1999 .

[7] B. Schölkopf,et al. Asymptotically Optimal Choice of ε-Loss for Support Vector Machines , 1998 .

[8] J. Weston,et al. Support vector regression with ANOVA decomposition kernels , 1999 .

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

[10] Bernhard Schölkopf,et al. New Support Vector Algorithms , 2000, Neural Computation.