Local Linear Approximation for Kernel Methods: The Railway Kernel
In this paper we present a new kernel, the Railway Kernel, that works properly for general (nonlinear) classification problems, with the interesting property that acts locally as a linear kernel. In this way, we avoid potential problems due to the use of a general purpose kernel, like the RBF kernel, as the high dimension of the induced feature space. As a consequence, following our methodology the number of support vectors is much lower and, therefore, the generalizacion capability of the proposed kernel is higher than the obtained using RBF kernels. Experimental work is shown to support the theoretical issues.
KeywordsSupport Vector Machine Kernel Method Linear Kernel Linear Support Vector Machine Local Linear Approximation
Unable to display preview. Download preview PDF.
- 2.Cristianini, N., Shawe-Taylor, J.: An introduction to Support Vector Machine. Cambridge University Press, Cambridge (2000)Google Scholar
- 5.Mangasarian, O.L., Wolberg, W.H.: Cancer diagnosis via linear programming. SIAM News 23(5), 1–18 (1990)Google Scholar
- 6.Moguerza, J., Muñoz, A.: SVM with aplications. In: Statistical Science ( in press, 2006)Google Scholar
- 9.Schölkopf, B., Herbrich, R., Smola, A., Williamson, R.: A Generalized Representer Theorem. NeuroCOLT2 TR Series, NC2-TR2000-81 (2000)Google Scholar
- 10.Schölkopf, B., Smola, A.: Learning with Kernels. MIT Press, Cambridge (2002)Google Scholar