How Robust is Federated Learning to Communication Error? A Comparison Study Between Uplink and Downlink Channels

被引:1
|
作者
Qu, Linping [1 ]
Song, Shenghui [1 ]
Tsui, Chi-Ying [1 ]
Mao, Yuyi [2 ]
机构
[1] Hong Kong Univ Sci & Technol, Dept ECE, Hong Kong, Peoples R China
[2] Hong Kong Polytech Univ, Dept EEE, Hong Kong, Peoples R China
来源
2024 IEEE WIRELESS COMMUNICATIONS AND NETWORKING CONFERENCE, WCNC 2024 | 2024年
关键词
Federated learning (FL); bit error rate (BER); uplink and downlink;
D O I
10.1109/WCNC57260.2024.10570687
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Because of its privacy-preserving capability, federated learning (FL) has attracted significant attention from both academia and industry. However, when being implemented over wireless networks, it is not clear how much communication error can be tolerated by FL. This paper investigates the robustness of FL to the uplink and downlink communication error. Our theoretical analysis reveals that the robustness depends on two critical parameters, namely the number of clients and the numerical range of model parameters. It is also shown that the uplink communication in FL can tolerate a higher bit error rate (BER) than downlink communication, and this difference is quantified by a proposed formula. The findings and theoretical analyses are further validated by extensive experiments.
引用
收藏
页数:6
相关论文
共 50 条
  • [31] Communication-Efficient and Byzantine-Robust Differentially Private Federated Learning
    Li, Min
    Xiao, Di
    Liang, Jia
    Huang, Hui
    IEEE COMMUNICATIONS LETTERS, 2022, 26 (08) : 1725 - 1729
  • [32] A comprehensive experimental comparison between federated and centralized learning
    Garst, Swier
    Dekker, Julian
    Reinders, Marcel
    DATABASE-THE JOURNAL OF BIOLOGICAL DATABASES AND CURATION, 2025, 2025
  • [33] A Comparison Study of Coupled and Decoupled Uplink-Downlink Access in Heterogeneous Cellular Networks
    Zhang, Lan
    Feng, Gang
    Qin, Shuang
    Nie, Weili
    2015 IEEE GLOBAL COMMUNICATIONS CONFERENCE (GLOBECOM), 2015,
  • [34] SparSFA: Towards robust and communication-efficient peer-to-peer federated learning
    Wang, Han
    Munoz-Gonzalez, Luis
    Hameed, Muhammad Zaid
    Eklund, David
    Raza, Shahid
    COMPUTERS & SECURITY, 2023, 129
  • [35] Performance comparison between the use of group signatures and robust error correction codes in wireless multiuser communication systems
    Tran, TA
    Sesay, AB
    IEEE CCEC 2002: CANADIAN CONFERENCE ON ELECTRCIAL AND COMPUTER ENGINEERING, VOLS 1-3, CONFERENCE PROCEEDINGS, 2002, : 1337 - 1341
  • [36] Uplink and Downlink Decoupled 5G/B5G Vehicular Networks: A Federated Learning Assisted Client Selection Method
    Liu, Tingting
    Zhou, Haibo
    Li, Jun
    Shu, Feng
    Han, Zhu
    IEEE TRANSACTIONS ON VEHICULAR TECHNOLOGY, 2023, 72 (02) : 2280 - 2292
  • [37] An Experimental Study of Byzantine-Robust Aggregation Schemes in Federated Learning
    Li, Shenghui
    Ngai, Edith
    Voigt, Thiemo
    IEEE TRANSACTIONS ON BIG DATA, 2024, 10 (06) : 975 - 988
  • [38] Learning How to Transfer From Uplink to Downlink via Hyper-Recurrent Neural Network for FDD Massive MIMO
    Liu, Yusha
    Simeone, Osvaldo
    IEEE TRANSACTIONS ON WIRELESS COMMUNICATIONS, 2022, 21 (10) : 7975 - 7989
  • [39] Communication-Efficient and Byzantine-Robust Federated Learning for Mobile Edge Computing Networks
    Zhang, Zhuangzhuang
    Wl, Libing
    He, Debiao
    Li, Jianxin
    Cao, Shuqin
    Wu, Xianfeng
    IEEE NETWORK, 2023, 37 (04): : 112 - 119
  • [40] Efficient Communication for Decentralized Federated Learning: An Energy Disaggregation Case Study
    Zhang, Yusen
    Gao, Feng
    Zhou, Kangjia
    IEEE 15TH INTERNATIONAL SYMPOSIUM ON POWER ELECTRONICS FOR DISTRIBUTED GENERATION SYSTEMS, PEDG 2024, 2024,