An efficient method for computing leave-one-out error in support vector machines with Gaussian kernels

Author

Lee, Martin M S ; Keerthi, S. Sathiya ; Ong, Chong Jin ; DeCoste, Dennis

Author_Institution

Dept. of Mech. Eng., Nat. Univ. of Singapore, Singapore

Volume

15

Issue

3

fYear

2004

fDate

5/1/2004 12:00:00 AM

Firstpage

750

Lastpage

757

Abstract

In this paper, we give an efficient method for computing the leave-one-out (LOO) error for support vector machines (SVMs) with Gaussian kernels quite accurately. It is particularly suitable for iterative decomposition methods of solving SVMs. The importance of various steps of the method is illustrated in detail by showing the performance on six benchmark datasets. The new method often leads to speedups of 10-50 times compared to standard LOO error computation. It has good promise for use in hyperparameter tuning and model comparison.

Keywords

Gaussian processes; iterative methods; support vector machines; Gaussian kernels; benchmark datasets; hyperparameter tuning; iterative decomposition methods; leave-one-out error computing; model comparison; support vector machines; Iterative methods; Kernel; Laboratories; Lagrangian functions; Learning systems; Machine learning; Mechanical engineering; Propulsion; Support vector machine classification; Support vector machines; Computing Methodologies; Normal Distribution; Research Design;

fLanguage

English

Journal_Title

Neural Networks, IEEE Transactions on

Publisher

ieee

ISSN

1045-9227

Type

jour

DOI

10.1109/TNN.2004.824266

Filename

1296700