Learning a Coupled Linearized Method in Online Setting

doi:10.1109/TNNLS.2016.2514413

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

	Learning a Coupled Linearized Method in Online Setting
	Xue, Wei1 ; Zhang, Wensheng1,2
发表期刊	IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
	2017-02-01
卷号	28 期号:2 页码:438-450
文章类型	Article
摘要	Based on the alternating direction method of multipliers, in this paper, we propose, analyze, and test a coupled linearized method, which aims to minimize an unconstrained problem consisting of a loss term and a regularization term in an online setting. To solve this problem, we first transform it into an equivalent constrained minimization problem with a separable structure. Then, we split the corresponding augmented Lagrangian function and minimize the resulting subproblems distributedly with one variable by fixing another one. This method is easy to execute without calculating matrix inversion by implementing three linearized operations per iteration, and at each iteration, we can obtain a closed-form solution. In particular, our update rule contains the well-known soft-thresholding operator as a special case. Moreover, upper bound on the regret of the proposed method is analyzed. Under some mild conditions, it can achieve 0(1/ root T) convergence rate for convex learning problems and 0((logT)/ T) for strongly convex learning. Numerical experiments and comparisons with several state-of-the-art methods are reported, which demonstrate the efficiency and effectiveness of our approach.
关键词	Alternating Minimization Convex Optimization Linearized Operation Online Learning Regret Bound
WOS标题词	Science & Technology ; Technology
DOI	10.1109/TNNLS.2016.2514413
关键词[WOS]	OPTIMIZATION METHODS ; ALGORITHM ; REGRESSION ; SHRINKAGE ; SELECTION
收录类别	SCI
语种	英语
项目资助者	National Natural Science Foundation of China(U1135005 ; Project of Post-Graduate Scientific Research Innovation Program of Jiangsu Province(KYZZ15_0123) ; 61305018 ; 61432008 ; 61472423 ; 61532006)
WOS研究方向	Computer Science ; Engineering
WOS类目	Computer Science, Artificial Intelligence ; Computer Science, Hardware & Architecture ; Computer Science, Theory & Methods ; Engineering, Electrical & Electronic
WOS记录号	WOS:000394522900017
引用统计	被引频次：3[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/173211/14383
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
作者单位	1.Nanjing Univ Sci & Technol, Sch Comp Sci & Engn, Nanjing 210094, Jiangsu, Peoples R China 2.Chinese Acad Sci, Inst Automat, Beijing 100190, Peoples R China
推荐引用方式 GB/T 7714	Xue, Wei,Zhang, Wensheng. Learning a Coupled Linearized Method in Online Setting[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,2017,28(2):438-450.
APA	Xue, Wei,&Zhang, Wensheng.(2017).Learning a Coupled Linearized Method in Online Setting.IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS,28(2),438-450.
MLA	Xue, Wei,et al."Learning a Coupled Linearized Method in Online Setting".IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 28.2(2017):438-450.