A FAST DUAL GRADIENT METHOD FOR SEPARABLE CONVEX OPTIMIZATION VIA SMOOTHING

被引：0

作者：

Li, Jueyou ^{[1
]}

Wu, Zhiyou ^{[1
]}

Wu, Changzhi ^{[2
]}

Long, Qiang ^{[3
]}

Wang, Xiangyu ^{[2
]}

Lee, Jae-Myung ^{[4
,5
,6
]}

Jung, Kwang-Hyo ^{[4
]}

机构：

[1] Chongqing Normal Univ, Sch Math Sci, Chongqing 400047, Peoples R China

[2] Curtin Univ, Sch Built Environm, Australasian Joint Res Ctr Bldg Informat Modellin, Perth, WA 6845, Australia

[3] Southwest Univ Sci & Technol, Sch Sci, Mianyang 621010, Sichuan, Peoples R China

[4] Pusan Natl Univ, Dept Naval Architecture & Ocean Engn, Busan, South Korea

[5] Pusan Natl Univ, Integrat Grad Program Ship & Offshore Plant Techn, BK21Plus, Busan, South Korea

[6] Pusan Natl Univ, Cryogen Mat Res Inst, Busan, South Korea

来源：

PACIFIC JOURNAL OF OPTIMIZATION | 2016年 / 12卷 / 02期

基金：

新加坡国家研究基金会; 澳大利亚研究理事会;

关键词：

convex optimization; dual decomposition; smoothing technique; fast gradient method; parallel computation; LAGRANGIAN DECOMPOSITION; NEURAL-NETWORKS; MINIMIZATION; STABILITY;

D O I：

暂无

中图分类号：

C93 [管理学]; O22 [运筹学];

学科分类号：

070105 ; 12 ; 1201 ; 1202 ; 120202 ;

摘要：

This paper considers a class of separable convex optimization problems with linear coupled constraints arising in many applications. Based on a novel smoothing technique, a simple fast dual gradient method is presented to solve this class of problems. Then the convergence of the proposed method is proved and the computational complexity bound of the method for achieving an approximately optimal solution is given explicitly. An improved iteration complexity bound is obtained when the objective function of the problem is strongly convex. Our algorithm is simple and fast, which can be implemented in a parallel fashion. Numerical experiments on a network utility maximization problem are presented to illustrate the effectiveness of the proposed algorithm.

引用

页码：289 / +

页数：17

共 30 条

[1] An O(1/k) Gradient Method for Network Resource Allocation Problems
Beck, Amir
Nedic, Angelia
Ozdaglar, Asuman
Teboulle, Marc
[J]. IEEE TRANSACTIONS ON CONTROL OF NETWORK SYSTEMS, 2014, 1 (01): : 64 - 73
[2] Bertsekas D. P., 1989, PARALLEL DISTRIBUTED, V23
[3] A double smoothing technique for solving unconstrained nondifferentiable convex optimization problems
Bot, Radu Ioan
Hendrich, Christopher
[J]. COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2013, 54 (02) : 239 - 262
[4] Distributed optimization and statistical learning via the alternating direction method of multipliers
Boyd S.
Parikh N.
Chu E.
Peleato B.
Eckstein J.
[J]. Foundations and Trends in Machine Learning, 2010, 3 (01): : 1 - 122
[5] A PROXIMAL-BASED DECOMPOSITION METHOD FOR CONVEX MINIMIZATION PROBLEMS
CHEN, G
TEBOULLE, M
[J]. MATHEMATICAL PROGRAMMING, 1994, 64 (01) : 81 - 101
[6] Connejo A., 2006, Decomposition Techniques_in_Mathematical_Programming:_Engineering_and_Science_Applications
[7] Multi-agent model predictive control of signaling split in urban traffic networks
de Oliveira, Lucas Barcelos
Camponogara, Eduardo
[J]. TRANSPORTATION RESEARCH PART C-EMERGING TECHNOLOGIES, 2010, 18 (01) : 120 - 139
[8] DOUBLE SMOOTHING TECHNIQUE FOR LARGE-SCALE LINEARLY CONSTRAINED CONVEX OPTIMIZATION
Devolder, Olivier
Glineur, Francois
Nesterov, Yurii
[J]. SIAM JOURNAL ON OPTIMIZATION, 2012, 22 (02) : 702 - 727
[9] State estimation distributed processing
Ebrahimian, R
Baldick, R
[J]. IEEE TRANSACTIONS ON POWER SYSTEMS, 2000, 15 (04) : 1240 - 1246
[10] Machine Learning Source Separation Using Maximum A Posteriori Nonnegative Matrix Factorization
Gao, Bin
Woo, W. L.
Ling, Bingo W-K.
[J]. IEEE TRANSACTIONS ON CYBERNETICS, 2014, 44 (07) : 1169 - 1179

← 1 2 3 →