Information criteria for support vector machines

Author

Kobayashi, K. ; Komaki, F.

Author_Institution

Inst. of Stat. Math., Tokyo

Volume

17

Issue

3

fYear

2006

fDate

5/1/2006 12:00:00 AM

Firstpage

571

Lastpage

577

Abstract

This paper presents kernel regularization information criterion (KRIC), which is a new criterion for tuning regularization parameters in kernel logistic regression (KLR) and support vector machines (SVMs). The main idea of the KRIC is based on the regularization information criterion (RIC). We derive an eigenvalue equation to calculate the KRIC and solve the problem. The computational cost for parameter tuning by the KRIC is reduced drastically by using the Nystroumlm approximation. The test error rate of SVMs or KLR with the regularization parameter tuned by the KRIC is comparable with the one by the cross validation or evaluation of the evidence. The computational cost of the KRIC is significantly lower than the one of the other criteria

Keywords

eigenvalues and eigenfunctions; regression analysis; support vector machines; Nystrom approximation; eigenvalue equation; kernel logistic regression; kernel regularization information criterion; parameter tuning; support vector machines; test error rate; Bayesian methods; Computational efficiency; Distribution functions; Eigenvalues and eigenfunctions; Equations; Error analysis; Kernel; Logistics; Support vector machines; Testing; Kernel logistic regression (KLR); kernel machine; parameter tuning; regularization information criterion; support vector machine (SVM); Algorithms; Artificial Intelligence; Computer Simulation; Information Storage and Retrieval; Models, Statistical; Pattern Recognition, Automated;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/TNN.2006.873276

Filename

1629082