FedHybrid: A Hybrid Federated Optimization Method for Heterogeneous Clients

被引:8
作者
Niu, Xiaochun [1 ]
Wei, Ermin [2 ]
机构
[1] Northwestern Univ, Dept Ind Engn & Management Sci, Evanston, IL 60208 USA
[2] Northwestern Univ, Dept Elect & Comp Engn, Dept Ind Engn & Management Sci, Evanston, IL 60208 USA
基金
美国国家科学基金会;
关键词
Optimization; Convergence; Servers; Topology; Newton method; Hardware; Convex functions; Server-client networks; distributed optimization; primal-dual algorithm; heterogeneous systems; DISTRIBUTED OPTIMIZATION; CONVERGENCE; FRAMEWORK;
D O I
10.1109/TSP.2023.3240083
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
We consider a distributed consensus optimization problem over a server-client (federated) network, where all clients are connected to a central server. Current distributed algorithms fail to capture the heterogeneity in clients' local computation capacities. Motivated by the method of multipliers in centralized optimization, we derive a Newton-type primal-dual method with a distributed implementation utilizing the server-client topology. Then we propose FedHybrid as a hybrid primal-dual method that allows heterogeneous clients to perform different types of updates. Specifically, those clients with higher computational capabilities and/or cheaper costs to perform computation can implement Newton-type updates locally, while other clients can adopt much simpler gradient-type updates. Theoretically, we propose a novel merit function by combining the dual optimality gap and the primal tracking error. We prove that FedHybrid converges linearly to the exact optimal point for strongly convex functions, regardless of clients' choices of gradient-type or Newton-type updates. Finally, we show numerical studies to demonstrate the efficacy of our method in practice. To the best of our knowledge, this is the first hybrid method allowing heterogeneous local updates for distributed consensus optimization with provable convergence and rate guarantees.
引用
收藏
页码:150 / 163
页数:14
相关论文
共 52 条
[1]  
Acar D. A. E., 2021, PROC INT C LEARN REP
[2]  
Arrow KJ., 1958, Studies in linear and nonlinear programming
[3]   A fast dual proximal gradient algorithm for convex minimization and applications [J].
Beck, Amir ;
Teboulle, Marc .
OPERATIONS RESEARCH LETTERS, 2014, 42 (01) :1-6
[4]   An investigation of Newton-Sketch and subsampled Newton methods [J].
Berahas, Albert S. ;
Bollapragada, Raghu ;
Nocedal, Jorge .
OPTIMIZATION METHODS & SOFTWARE, 2020, 35 (04) :661-680
[5]  
Bertsekas D., 1989, Parallel and distributed computation: numerical methods
[6]  
Bertsekas D., 1982, CONSTRAINED OPTIMIZA
[7]   Distributed optimization and statistical learning via the alternating direction method of multipliers [J].
Boyd S. ;
Parikh N. ;
Chu E. ;
Peleato B. ;
Eckstein J. .
Foundations and Trends in Machine Learning, 2010, 3 (01) :1-122
[8]  
Boyd S., 2004, CONVEX OPTIMIZATION, DOI [DOI 10.1017/CBO9780511804441, 10.1017/CBO9780511804441]
[10]   An Overview of Recent Progress in the Study of Distributed Multi-Agent Coordination [J].
Cao, Yongcan ;
Yu, Wenwu ;
Ren, Wei ;
Chen, Guanrong .
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2013, 9 (01) :427-438