Distributed Mirror Descent Algorithm With Bregman Damping for Nonsmooth Constrained Optimization

被引:7
|
作者
Chen, Guanpu [1 ]
Xu, Gehui [1 ]
Li, Weijian [2 ]
Hong, Yiguang [3 ,4 ]
机构
[1] Chinese Acad Sci, Acad Math & Syst Sci, Key Lab Syst & Control, Beijing 100190, Peoples R China
[2] DAMO Acad, Alibaba Grp, Decis Intelligence Lab, Hangzhou 310052, Peoples R China
[3] Tongji Univ, Dept Control Sci & Engn, Shanghai 210201, Peoples R China
[4] Tongji Univ, Shanghai Res Inst Intelligent Autonomous Syst, Shanghai 210201, Peoples R China
基金
中国国家自然科学基金;
关键词
Constrained optimization; distributed algorithm; mirror descent; multi-agent system; nonsmooth; CONVERGENCE ANALYSIS; ECONOMIC-DISPATCH; CONSENSUS;
D O I
10.1109/TAC.2023.3244995
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
To efficiently solve the nonsmooth distributed optimization with both local constraints and coupled constraints, we propose a distributed continuous-time algorithm based on the mirror descent (MD) method. In this article, we introduce the Bregman damping into distributed MD-based dynamics, which not only successfully applies the MD idea to the distributed primal-dual framework, but also ensures the boundedness of all variables and the convergence of the entire dynamics. Our approach generalizes the classic distributed projection-based dynamics, and establishes a connection between MD methods and distributed Euclidean-projected approaches. Also, we prove the convergence of the proposed distributed dynamics with an O(1/t) rate. For practical implementation, we further give a discrete-time algorithm based on the proposed dynamics with an O(1/root k) convergence rate.
引用
收藏
页码:6921 / 6928
页数:8
相关论文
共 50 条
  • [41] A decomposition algorithm for a class of constrained nonsmooth convex optimization problems
    Li, Dan
    Chen, Shuang
    Meng, Fan-Yun
    2016 IEEE INFORMATION TECHNOLOGY, NETWORKING, ELECTRONIC AND AUTOMATION CONTROL CONFERENCE (ITNEC), 2016, : 685 - 690
  • [42] Block-Coordinate Gradient Descent Method for Linearly Constrained Nonsmooth Separable Optimization
    P. Tseng
    S. Yun
    Journal of Optimization Theory and Applications, 2009, 140
  • [43] Block-Coordinate Gradient Descent Method for Linearly Constrained Nonsmooth Separable Optimization
    Tseng, P.
    Yun, S.
    JOURNAL OF OPTIMIZATION THEORY AND APPLICATIONS, 2009, 140 (03) : 513 - 535
  • [44] Distributed Resilient Initialization-Free Jacobi Descent Algorithm for Constrained Optimization Against DoS Attacks
    Li, Yushuai
    Huang, Bonan
    Dai, Jing
    Gao, David Wenzhong
    Sun, Qiuye
    Zhang, Huaguang
    IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING, 2024, 21 (03) : 3332 - 3343
  • [45] Taming Nonconvex Stochastic Mirror Descent with General Bregman Divergence
    Fatkhullin, Ilyas
    He, Niao
    INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE AND STATISTICS, VOL 238, 2024, 238
  • [46] On the efficiency of a randomized mirror descent algorithm in online optimization problems
    A. V. Gasnikov
    Yu. E. Nesterov
    V. G. Spokoiny
    Computational Mathematics and Mathematical Physics, 2015, 55 : 580 - 596
  • [47] On the efficiency of a randomized mirror descent algorithm in online optimization problems
    Gasnikov, A. V.
    Nesterov, Yu. E.
    Spokoiny, V. G.
    COMPUTATIONAL MATHEMATICS AND MATHEMATICAL PHYSICS, 2015, 55 (04) : 580 - 596
  • [48] Stochastic mirror descent method for distributed multi-agent optimization
    Jueyou Li
    Guoquan Li
    Zhiyou Wu
    Changzhi Wu
    Optimization Letters, 2018, 12 : 1179 - 1197
  • [49] Stochastic mirror descent method for distributed multi-agent optimization
    Li, Jueyou
    Li, Guoquan
    Wu, Zhiyou
    Wu, Changzhi
    OPTIMIZATION LETTERS, 2018, 12 (06) : 1179 - 1197
  • [50] Adaptive quantized online distributed mirror descent algorithm with Bandit feedback
    Xie J.-R.
    Gao W.-H.
    Xie Y.-B.
    Kongzhi Lilun Yu Yingyong/Control Theory and Applications, 2023, 40 (10): : 1774 - 1782