Support vector machines regression with l1-regularizer

JOURNAL OF APPROXIMATION THEORY

卷:164

Article

Tong, Hongzhi¹  Chen, Di-Rong^2,3  Yang, Fenghong⁴ 

[1] Univ Int Business & Econ, Sch Informat Technol & Management, Beijing 100029, Peoples R China

[2] Beijing Univ Aeronaut & Astronaut, Dept Math, Beijing 100083, Peoples R China

[3] Beijing Univ Aeronaut & Astronaut, LMIB, Beijing 100083, Peoples R China

[4] Cent Univ Finance & Econ, Sch Appl Math, Beijing 100081, Peoples R China

关键词: Support vector machines regression;    Coefficient regularization;    Learning rate;    Reproducing kernel Hilbert spaces;    Error decomposition;   

DOI  :  10.1016/j.jat.2012.06.005

来源: Elsevier

【 摘 要 】

The classical support vector machines regression (SVMR) is known as a regularized learning algorithm in reproducing kernel Hilbert spaces (RKHS) with a epsilon-insensitive loss function and an RKHS norm regularizer. In this paper, we study a new SVMR algorithm where the regularization term is proportional to l(1)-norm of the coefficients in the kernel ensembles. We provide an error analysis of this algorithm, an explicit learning rate is then derived under some assumptions. (c) 2012 Elsevier Inc. All rights reserved.

【 授权许可】

Free