Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss

The problem of learning the kernel function with linear combinations of multiple kernels has attracted considerable attention recently in machine learning. Specially, by imposing an lp-norm penalty on the kernel combination coefficient, multiple kernel learning (MKL) was proved useful and effective...

Full description

Saved in:
Bibliographic Details
Main Authors: Shao-Gao Lv, Jin-De Zhu
Format: Article
Language:English
Published: Wiley 2012-01-01
Series:Abstract and Applied Analysis
Online Access:http://dx.doi.org/10.1155/2012/915920
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850220103290847232
author Shao-Gao Lv
Jin-De Zhu
author_facet Shao-Gao Lv
Jin-De Zhu
author_sort Shao-Gao Lv
collection DOAJ
description The problem of learning the kernel function with linear combinations of multiple kernels has attracted considerable attention recently in machine learning. Specially, by imposing an lp-norm penalty on the kernel combination coefficient, multiple kernel learning (MKL) was proved useful and effective for theoretical analysis and practical applications (Kloft et al., 2009, 2011). In this paper, we present a theoretical analysis on the approximation error and learning ability of the lp-norm MKL. Our analysis shows explicit learning rates for lp-norm MKL and demonstrates some notable advantages compared with traditional kernel-based learning algorithms where the kernel is fixed.
format Article
id doaj-art-7790ca5ad25245fb876b8c5e94dc0cff
institution OA Journals
issn 1085-3375
1687-0409
language English
publishDate 2012-01-01
publisher Wiley
record_format Article
series Abstract and Applied Analysis
spelling doaj-art-7790ca5ad25245fb876b8c5e94dc0cff2025-08-20T02:07:10ZengWileyAbstract and Applied Analysis1085-33751687-04092012-01-01201210.1155/2012/915920915920Error Bounds for lp-Norm Multiple Kernel Learning with Least Square LossShao-Gao Lv0Jin-De Zhu1Statistics School, Southwestern University of Finance and Economics, Chengdu 611130, ChinaThe 2nd Geological Party of Bureau of Geology and Mineral Resources, Henan, Jiaozuo 450000, ChinaThe problem of learning the kernel function with linear combinations of multiple kernels has attracted considerable attention recently in machine learning. Specially, by imposing an lp-norm penalty on the kernel combination coefficient, multiple kernel learning (MKL) was proved useful and effective for theoretical analysis and practical applications (Kloft et al., 2009, 2011). In this paper, we present a theoretical analysis on the approximation error and learning ability of the lp-norm MKL. Our analysis shows explicit learning rates for lp-norm MKL and demonstrates some notable advantages compared with traditional kernel-based learning algorithms where the kernel is fixed.http://dx.doi.org/10.1155/2012/915920
spellingShingle Shao-Gao Lv
Jin-De Zhu
Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss
Abstract and Applied Analysis
title Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss
title_full Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss
title_fullStr Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss
title_full_unstemmed Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss
title_short Error Bounds for lp-Norm Multiple Kernel Learning with Least Square Loss
title_sort error bounds for lp norm multiple kernel learning with least square loss
url http://dx.doi.org/10.1155/2012/915920
work_keys_str_mv AT shaogaolv errorboundsforlpnormmultiplekernellearningwithleastsquareloss
AT jindezhu errorboundsforlpnormmultiplekernellearningwithleastsquareloss