Distributed Gradient Tracking for Unbalanced Optimization With Different Constraint Sets

被引:26
作者
Cheng, Songsong [1 ]
Liang, Shu [2 ]
Fan, Yuan [1 ]
Hong, Yiguang [2 ]
机构
[1] Anhui Univ, Sch Elect Engn & Automat, Hefei 230601, Peoples R China
[2] Tongji Univ, Dept Control Sci & Engn, Shanghai 200092, Peoples R China
基金
中国国家自然科学基金;
关键词
Optimization; Convergence; Directed graphs; Convex functions; Multi-agent systems; Linear programming; Heuristic algorithms; different constraint sets; distrib- uted optimization; gradient tracking; unbalanced graphs; ALGORITHM; CONVERGENCE; CONSENSUS;
D O I
10.1109/TAC.2022.3192316
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
tracking methods have become popular for distributed optimization in recent years, partially because they achieve linear convergence using only a constant step-size for strongly convex optimization. In this article, we construct a counterexample on constrained optimization to show that direct extension of gradient tracking by using projections cannot guarantee the correctness. Then, we propose projected gradient tracking algorithms with diminishing step-sizes rather than a constant one for distributed strongly convex optimization with different constraint sets and unbalanced graphs. Our basic algorithm can achieve O(ln T/T ) convergence rate. Moreover, we design an epoch iteration scheme and improve the convergence rate as O(1/T ).
引用
收藏
页码:3633 / 3640
页数:8
相关论文
共 25 条
[1]   Decentralized Proximal Gradient Algorithms With Linear Convergence Rates [J].
Alghunaim, Sulaiman A. ;
Ryu, Ernest K. ;
Yuan, Kun ;
Sayed, Ali H. .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2021, 66 (06) :2787-2794
[2]   DISTRIBUTED OPTIMIZATION FOR MULTI-AGENT SYSTEM OVER UNBALANCED GRAPHS WITH LINEAR CONVERGENCE RATE [J].
Cheng, Songsong ;
Liang, Shu .
KYBERNETIKA, 2020, 56 (03) :559-577
[3]   Distributed Continuous-Time Convex Optimization on Weight-Balanced Digraphs [J].
Gharesifard, Bahman ;
Cortes, Jorge .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2014, 59 (03) :781-786
[4]   Distributed Convex Optimization on State-Dependent Undirected Graphs: Homogeneity Technique [J].
Hong, Huifen ;
Yu, Xinghuo ;
Yu, Wenwu ;
Zhang, Dong ;
Wen, Guanghui .
IEEE TRANSACTIONS ON CONTROL OF NETWORK SYSTEMS, 2020, 7 (01) :42-52
[5]   Linear Convergence of Consensus-Based Quantized Optimization for Smooth and Strongly Convex Cost Functions [J].
Kajiyama, Yuichi ;
Hayashi, Naoki ;
Takai, Shigemasa .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2021, 66 (03) :1254-1261
[6]   Primal-dual algorithm for distributed constrained optimization [J].
Lei, Jinlong ;
Chen, Han-Fu ;
Fang, Hai-Tao .
SYSTEMS & CONTROL LETTERS, 2016, 96 :110-117
[7]  
Li BY, 2020, J MACH LEARN RES, V21
[8]   Dual Averaging Push for Distributed Convex Optimization Over Time-Varying Directed Graph [J].
Liang, Shu ;
Wang, Le Yi ;
Yin, George .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2020, 65 (04) :1785-1791
[9]   Distributed Smooth Convex Optimization With Coupled Constraints [J].
Liang, Shu ;
Wang, Le Yi ;
Yin, George .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2020, 65 (01) :347-353
[10]   Online Distributed Optimization With Strongly Pseudoconvex-Sum Cost Functions [J].
Lu, Kaihong ;
Jing, Gangshan ;
Wang, Long .
IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 2020, 65 (01) :426-433