Recursive Finite Newton Algorithm

for Support Vector Regression in the Primal

Liefeng Bo, Ling Wang, and Licheng Jiao

Xidian University

Abstract: Some algorithms in the primal have been recently proposed for training support vector machines. This letter follows those studies and develops a recursive finite Newton algorithm (IHLF-SVR-RFN) [1] for training nonlinear support vector regression [2]. The insensitive Huber loss function and the computation of the Newton step are discussed in detail. Comparisons with LIBSVM 2.82 show that the proposed algorithm gives promising results.

References

  1. Liefeng Bo, Ling Wang, and Licheng Jiao, Recursive Finite Newton Algorithm for Support Vector Regression in the Primal, Neural Computation (NECO), vol. 19(4), pp. 1082-1096, 2007. [PDF] [BIB]

  2. A. Smola and B. Scholkopf, A Tutorial on Support Vector Regression, Statistics and Computing, vol. 14,  pp. 199 - 222, 2004..

Matlab Source Code

Description: IHLF-SVR-RFN is  a package for training support vector regression using the recursive finite Newton algorithm.

Requirement: Matlab 7.01.

Download: [code]. This package is free for academic usage. You can run it at your own risk.