Fast Leave-One-Out Evaluation and Improvement on Inference for LS-SVMs

Authors:
Zhao Ying;Kwoh Chee Keong
Affiliations:
Nanyang Technological University, Singapore;Nanyang Technological University, Singapore
Venue:
ICPR '04 Proceedings of the Pattern Recognition, 17th International Conference on (ICPR'04) Volume 3 - Volume 03
Year:
2004

Citing 0
Cited 4

Fast cross-validation algorithms for least squares support vector machine and kernel ridge regression

Pattern Recognition
Recursive reduced least squares support vector regression

Pattern Recognition
Efficient Optimization of the Parameters of LS-SVM for Regression versus Cross-Validation Error

ICANN '09 Proceedings of the 19th International Conference on Artificial Neural Networks: Part II
Optimized fixed-size kernel models for large data sets

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

In this paper, a fast leave-one-out (LOO) evaluation formula is introduced for least squares support vector machine (LS-SVM) classifiers. The computation cost can be reduced to approximately 1/N when compared to normal LOO procedure (is the number of training samples). Inspired by its fast speed, we are able to use it to replace the original Level 3 posterior probability approximation formula of the Bayesian framework [Bayesian framework for least squares support vector machine classifiers, gaussian processes and kernel fisher discriminant analysis] for LS-SVM classifiers. The improved inference framework shows higher generalization performance and faster computation speed.