A general soft method for learning SVM classifiers with L-1-norm penalty

doi:10.1016/j.patcog.2007.08.004

	A general soft method for learning SVM classifiers with L-1-norm penalty
	Tao, Qing ; Wu, Gao-Wei ; Wang, Jue
	2008-03-01
发表期刊	PATTERN RECOGNITION
ISSN	0031-3203
卷号	41 期号:3 页码:939-948
摘要	Based on the geometric interpretation of support vector machines (SVMs), this paper presents a general technique that allows almost all the existing L-2-norm penalty based geometric algorithms, including Gilbert's algorithm, Schlesinger-Kozinec's (SK) algorithm and Mitchell-Dem'yanov-Malozemov's (MDM) algorithm, to be softened to achieve the corresponding learning L-1-SVM classifiers. Intrinsically, the resulting soft algorithms are to find E-optimal nearest points between two soft convex hulls. Theoretical analysis has indicated that our proposed soft algorithms are essentially generalizations of the corresponding existing hard algorithms, and consequently, they have the same properties of convergence and almost the identical cost of computation. As a specific example, the problem of solving nu-SVMs by the proposed soft MDM algorithm is investigated and the corresponding solution procedure is specified and analyzed. To validate the general soft technique, several real classification experiments are conducted with the proposed L-1-norm based MDM algorithms and numerical results have demonstrated that their performance is competitive to that of the corresponding L-2-norm based algorithms, such as SK and MDM algorithms. (C) 2007 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
关键词	support vector machines classification nu-SVMs nearest points Gilbert's algorithms Schlesinger-Kozinec's algorithms Mitchell-Dem'yanov-Malozemov's algorithms soft convex hulls
DOI	10.1016/j.patcog.2007.08.004
收录类别	SCI
语种	英语
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Engineering, Electrical & Electronic
WOS记录号	WOS:000251357100015
出版者	PERGAMON-ELSEVIER SCIENCE LTD
引用统计	被引频次：29[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://119.78.100.204/handle/2XEOYT63/11243
专题	中国科学院计算技术研究所期刊论文_英文
通讯作者	Tao, Qing
作者单位	1.Chinese Acad Sci, Inst Automat, Lab Complex Syst & Intelligence Sci, Beijing 100080, Peoples R China 2.New Star Res Inst Appl Tech, Hefei 230031, Peoples R China 3.Chinese Acad Sci, Comp Technol Inst, Div Intelligent Software Syst, Beijing 100080, Peoples R China
推荐引用方式 GB/T 7714	Tao, Qing,Wu, Gao-Wei,Wang, Jue. A general soft method for learning SVM classifiers with L-1-norm penalty[J]. PATTERN RECOGNITION,2008,41(3):939-948.
APA	Tao, Qing,Wu, Gao-Wei,&Wang, Jue.(2008).A general soft method for learning SVM classifiers with L-1-norm penalty.PATTERN RECOGNITION,41(3),939-948.
MLA	Tao, Qing,et al."A general soft method for learning SVM classifiers with L-1-norm penalty".PATTERN RECOGNITION 41.3(2008):939-948.