Optimization-Based GenQSGD for Federated Edge Learning

被引:0
|
作者
Li, Yangchen [1 ]
Cui, Ying [1 ]
Lau, Vincent [2 ]
机构
[1] Shanghai Jiao Tong Univ, Shanghai, Peoples R China
[2] HKUST, Hong Kong, Peoples R China
基金
上海市自然科学基金;
关键词
Federated learning; stochastic gradient descent; optimization; algorithm design; convergence analysis;
D O I
10.1109/GLOBECOM46510.2021.9685591
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
Optimal algorithm design for federated learning (FL) remains an open problem. This paper explores the full potential of FL in practical edge computing systems where workers may have different computation and communication capabilities, and quantized intermediate model updates are sent between the server and workers. First, we present a general quantized parallel mini-batch stochastic gradient descent (SGD) algorithm for FL, namely GenQSGD, which is parameterized by the number of global iterations, the numbers of local iterations at all workers, and the mini-batch size. We also analyze its convergence error for any choice of the algorithm parameters. Then, we optimize the algorithm parameters to minimize the energy cost under the time constraint and convergence error constraint. The optimization problem is a challenging non-convex problem with non-differentiable constraint functions. We propose an iterative algorithm to obtain a KKT point using advanced optimization techniques. Numerical results demonstrate the significant gains of GenQSGD over existing FL algorithms and reveal the importance of optimally designing FL algorithms.
引用
收藏
页数:6
相关论文
共 50 条
  • [21] Edge assignment in edge federated learning
    Thuy Do
    Duc A. Tran
    Anh Vo
    SN Applied Sciences, 2023, 5
  • [22] Snake swarm optimization-based deep reinforcement learning for resource allocation in edge computing environment
    Kaliraj, S.
    Sivakumar, V.
    Premkumar, N.
    Vatchala, S.
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2024, 36 (18):
  • [23] Lyapunov-Based Optimization of Edge Resources for Energy-Efficient Adaptive Federated Learning
    Battiloro, Claudio
    Di Lorenzo, Paolo
    Merluzzi, Mattia
    Barbarossa, Sergio
    IEEE TRANSACTIONS ON GREEN COMMUNICATIONS AND NETWORKING, 2023, 7 (01): : 265 - 280
  • [24] EdgeFed: Optimized Federated Learning Based on Edge Computing
    Ye, Yunfan
    Li, Shen
    Liu, Fang
    Tang, Yonghao
    Hu, Wanting
    IEEE ACCESS, 2020, 8 : 209191 - 209198
  • [25] Edge Computing Based on Federated Learning for Machine Monitoring
    Tsai, Yao-Hong
    Chang, Dong-Meau
    Hsu, Tse-Chuan
    APPLIED SCIENCES-BASEL, 2022, 12 (10):
  • [26] Optimization-Based Constrained Iterative Learning Control
    Mishra, Sandipan
    Topcu, Ufuk
    Tomizuka, Masayoshi
    IEEE TRANSACTIONS ON CONTROL SYSTEMS TECHNOLOGY, 2011, 19 (06) : 1613 - 1621
  • [27] Pareto Optimization-Based Iterative Learning Control
    Lim, Ingyu
    Barton, Kira L.
    2013 AMERICAN CONTROL CONFERENCE (ACC), 2013, : 5171 - 5176
  • [28] Vehicle Selection and Resource Optimization for Federated Learning in Vehicular Edge Computing
    Xiao, Huizi
    Zhao, Jun
    Pei, Qingqi
    Feng, Jie
    Liu, Lei
    Shi, Weisong
    IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS, 2022, 23 (08) : 11073 - 11087
  • [29] Fast-Convergence Federated Edge Learning via Bilevel Optimization
    Wang, Sai
    Gong, Yi
    2023 28TH ASIA PACIFIC CONFERENCE ON COMMUNICATIONS, APCC 2023, 2023, : 87 - 92
  • [30] DYNAMIC RESOURCE OPTIMIZATION FOR ADAPTIVE FEDERATED LEARNING AT THE WIRELESS NETWORK EDGE
    Di Lorenzo, Paolo
    Battiloro, Claudio
    Merluzzi, Mattia
    Barbarossa, Sergio
    2021 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH AND SIGNAL PROCESSING (ICASSP 2021), 2021, : 4910 - 4914