The direct extension of ADMM for multi-block convex minimization problems is not necessarily convergent

被引:20
|
作者
Caihua Chen
Bingsheng He
Yinyu Ye
Xiaoming Yuan
机构
[1] Nanjing University,International Centre of Management Science and Engineering, School of Management and Engineering
[2] Nanjing University,Department of Mathematics
[3] Stanford University,Department of Management Science and Engineering, School of Engineering
[4] Hong Kong Baptist University,Department of Mathematics
来源
Mathematical Programming | 2016年 / 155卷
关键词
Alternating direction method of multipliers; Convergence analysis; Convex programming; Splitting methods; 90C25; 90C30; 65K13;
D O I
暂无
中图分类号
学科分类号
摘要
The alternating direction method of multipliers (ADMM) is now widely used in many fields, and its convergence was proved when two blocks of variables are alternatively updated. It is strongly desirable and practically valuable to extend the ADMM directly to the case of a multi-block convex minimization problem where its objective function is the sum of more than two separable convex functions. However, the convergence of this extension has been missing for a long time—neither an affirmative convergence proof nor an example showing its divergence is known in the literature. In this paper we give a negative answer to this long-standing open question: The direct extension of ADMM is not necessarily convergent. We present a sufficient condition to ensure the convergence of the direct extension of ADMM, and give an example to show its divergence.
引用
收藏
页码:57 / 79
页数:22
相关论文
共 35 条
  • [1] The direct extension of ADMM for multi-block convex minimization problems is not necessarily convergent
    Chen, Caihua
    He, Bingsheng
    Ye, Yinyu
    Yuan, Xiaoming
    MATHEMATICAL PROGRAMMING, 2016, 155 (1-2) : 57 - 79
  • [2] Convergence analysis of the direct extension of ADMM for multiple-block separable convex minimization
    Tao, Min
    Yuan, Xiaoming
    ADVANCES IN COMPUTATIONAL MATHEMATICS, 2018, 44 (03) : 773 - 813
  • [3] Convergence analysis of the direct extension of ADMM for multiple-block separable convex minimization
    Min Tao
    Xiaoming Yuan
    Advances in Computational Mathematics, 2018, 44 : 773 - 813
  • [4] Convergent prediction-correction-based ADMM for multi-block separable convex programming
    Chang, Xiaokai
    Liu, Sanyang
    Zhao, Pengjun
    Li, Xu
    JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, 2018, 335 : 270 - 288
  • [5] On the convergence of the direct extension of ADMM for three-block separable convex minimization models with one strongly convex function
    Cai, Xingju
    Han, Deren
    Yuan, Xiaoming
    COMPUTATIONAL OPTIMIZATION AND APPLICATIONS, 2017, 66 (01) : 39 - 73
  • [6] On the convergence of the direct extension of ADMM for three-block separable convex minimization models with one strongly convex function
    Xingju Cai
    Deren Han
    Xiaoming Yuan
    Computational Optimization and Applications, 2017, 66 : 39 - 73
  • [7] A Convergent 3-Block Semi-Proximal ADMM for Convex Minimization Problems with One Strongly Convex Block
    Li, Min
    Sun, Defeng
    Toh, Kim-Chuan
    ASIA-PACIFIC JOURNAL OF OPERATIONAL RESEARCH, 2015, 32 (04)
  • [8] Convergence Analysis of L-ADMM for Multi-block Linear-Constrained Separable Convex Minimization Problem
    Feng J.-K.
    Zhang H.-B.
    Cheng C.-Z.
    Pei H.-M.
    J. Oper. Res. Soc. China, 4 (563-579): : 563 - 579
  • [9] A distributed Douglas-Rachford splitting method for multi-block convex minimization problems
    He, Hongjin
    Han, Deren
    ADVANCES IN COMPUTATIONAL MATHEMATICS, 2016, 42 (01) : 27 - 53
  • [10] A distributed Douglas-Rachford splitting method for multi-block convex minimization problems
    Hongjin He
    Deren Han
    Advances in Computational Mathematics, 2016, 42 : 27 - 53