Kernel Optimization Using a Generalized Eigenvalue Approach
There is no single generic kernel that suits all estimation tasks. Kernels that are learnt from the data are known to yield better classification. The coefficients of the optimal kernel that maximizes the class separability in the empirical feature space had been previously obtained by a gradient-based procedure. In this paper, we show how these coefficients can be learnt from the data by simply solving a generalized eigenvalue problem. Our approach yields a significant reduction in classification errors on selected UCI benchmarks.
KeywordsData dependent kernel Fisher’s coefficient generalized eigenvalue kernel optimization Rayleigh quotient
- 4.Murphy, P.M., Aha, D.W.: UCI machine learning repository (1992), http://www.ics.uci.edu/~mlearn/MLRepository.html
- 5.Jaakkola, T.S., Haussler, D.: Exploiting generating models in discriminative classifiers. In: Proc. of Tenth Conference on Advances in Neural Information Processing Systems, Denver (December 1998)Google Scholar
- 8.GEVP, Robust Control Toolbox. The Mathworks, http://www.mathworks.com/access/helpdesk/help/toolbox/robust/gevp.html