Sample-based online learning for bi-regular hinge loss

doi:10.1007/s13042-020-01272-7

CASIA OpenIR > 多模态人工智能系统全国重点实验室 > 人工智能与机器学习（杨雪冰）-技术团队

	Sample-based online learning for bi-regular hinge loss
	Xue, Wei1,2,3 ; Zhong, Ping 1; Zhang, Wensheng4 ; Yu, Gaohang 5; Chen, Yebin 2
发表期刊	INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS
ISSN	1868-8071
	2021-01-24
页码	16
通讯作者	Xue, Wei(cswxue@ahut.edu.cn) ; Zhong, Ping(zhongping@nudt.edu.cn)
摘要	Support vector machine (SVM), a state-of-the-art classifier for supervised classification task, is famous for its strong generalization guarantees derived from the max-margin property. In this paper, we focus on the maximum margin classification problem cast by SVM and study the bi-regular hinge loss model, which not only performs feature selection but tends to select highly correlated features together. To solve this model, we propose an online learning algorithm that aims at solving a non-smooth minimization problem by alternating iterative mechanism. Basically, the proposed algorithm alternates between intrusion samples detection and iterative optimization, and at each iteration it obtains a closed-form solution to the model. In theory, we prove that the proposed algorithm achieves O(1/root T) convergence rate under some mild conditions, where T is the number of training samples received in online learning. Experimental results on synthetic data and benchmark datasets demonstrate the effectiveness and performance of our approach in comparison with several popular algorithms, such as LIBSVM, SGD, PEGASOS, SVRG, etc.
关键词	SVM Max-margin classification Hinge loss Elastic net Online learning
DOI	10.1007/s13042-020-01272-7
收录类别	SCI
语种	英语
资助项目	National Natural Science Foundation of China[12071104] ; National Natural Science Foundation of China[61671456] ; National Natural Science Foundation of China[61806004] ; National Natural Science Foundation of China[61971428] ; China Postdoctoral Science Foundation[2020T130767] ; Natural Science Foundation of the Anhui Higher Education Institutions of China[KJ2019A0082] ; Natural Science Foundation of Zhejiang Province, China[LD19A010002]
项目资助者	National Natural Science Foundation of China ; China Postdoctoral Science Foundation ; Natural Science Foundation of the Anhui Higher Education Institutions of China ; Natural Science Foundation of Zhejiang Province, China
WOS研究方向	Computer Science
WOS类目	Computer Science, Artificial Intelligence
WOS记录号	WOS:000610860500001
出版者	SPRINGER HEIDELBERG
引用统计	被引频次：1[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.ia.ac.cn/handle/173211/42881
专题	多模态人工智能系统全国重点实验室_人工智能与机器学习（杨雪冰）-技术团队
通讯作者	Xue, Wei; Zhong, Ping
作者单位	1.Natl Univ Defense Technol, Natl Key Lab Sci & Technol Automatic Target Recog, Changsha 410073, Peoples R China 2.Anhui Univ Technol, Sch Comp Sci & Technol, Maanshan 243032, Peoples R China 3.Southeast Univ, Key Lab Comp Network & Informat Integrat, Minist Educ, Nanjing 211189, Peoples R China 4.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China 5.Hangzhou Dianzi Univ, Sch Sci, Hangzhou 310018, Peoples R China
推荐引用方式 GB/T 7714	Xue, Wei,Zhong, Ping,Zhang, Wensheng,et al. Sample-based online learning for bi-regular hinge loss[J]. INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS,2021:16.
APA	Xue, Wei,Zhong, Ping,Zhang, Wensheng,Yu, Gaohang,&Chen, Yebin.(2021).Sample-based online learning for bi-regular hinge loss.INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS,16.
MLA	Xue, Wei,et al."Sample-based online learning for bi-regular hinge loss".INTERNATIONAL JOURNAL OF MACHINE LEARNING AND CYBERNETICS (2021):16.