Distributed Momentum-Based Frank-Wolfe Algorithm for Stochastic Optimization

作     者：Jie Hou Xianlin Zeng Gang Wang Jian Sun Jie Chen Jie Hou;Xianlin Zeng;Gang Wang;Jian Sun;Jie Chen

作者机构：National Key Laboratory of Autonomous Intelligent Unmanned SystemsSchool of AutomationBeijing Institute of TechnologyBeijing 100081China National Key Laboratory of Autonomous Intelligent Unmanned SystemsSchool of AutomationBeijing Institute of TechnologyBeijing 100081 Beijing Institute of Technology Chongqing Innovation CenterChongqing 401120China School of Electronic and Information EngineeringTongji UniversityShanghai 200082and also with the National Key Laboratory of Autonomous Intelligent Unmanned Systemsthe School of AutomationBeijing Institute of TechnologyBeijing 100081China IEEE 

出 版 物：《IEEE/CAA Journal of Automatica Sinica》 (自动化学报（英文版）)

年 卷 期：2023年第10卷第3期

页      面：685-699页

核心收录：

学科分类：0711[理学-系统科学] 12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：supported in part by the National Key R&D Program of China(2021YFB1714800) the National Natural Science Foundation of China(62222303,62073035,62173034,61925303,62088101,61873033) the CAAI-Huawei MindSpore Open Fund the Chongqing Natural Science Foundation(2021ZX4100027) 

主　　题：Distributed optimization Frank-Wolfe(FW)algorithms momentum-based method stochastic optimization 

摘      要：This paper considers distributed stochastic optimization,in which a number of agents cooperate to optimize a global objective function through local computations and information exchanges with neighbors over a *** optimization problems are usually tackled by variants of projected stochastic gradient ***,projecting a point onto a feasible set is often *** Frank-Wolfe(FW)method has well-documented merits in handling convex constraints,but existing stochastic FW algorithms are basically developed for centralized *** this context,the present work puts forth a distributed stochastic Frank-Wolfe solver,by judiciously combining Nesterov s momentum and gradient tracking techniques for stochastic convex and nonconvex optimization over *** is shown that the convergence rate of the proposed algorithm is O(k^(-1/2))for convex optimization,and O(1/log_(2)(k))for nonconvex *** efficacy of the algorithm is demonstrated by numerical simulations against a number of competing alternatives.