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
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
  • [1] Communication-Efficient Nonconvex Federated Learning With Error Feedback for Uplink and Downlink
    Zhou, Xingcai
    Chang, Le
    Cao, Jinde
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2025, 36 (01) : 1003 - 1014
  • [2] Communication-Efficient Nonconvex Federated Learning With Error Feedback for Uplink and Downlink
    Zhou, Xingcai
    Chang, Le
    Cao, Jinde
    IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 2025, 36 (01) : 1003 - 1014
  • [3] Design and Analysis of Uplink and Downlink Communications for Federated Learning
    Zheng, Sihui
    Shen, Cong
    Chen, Xiang
    IEEE JOURNAL ON SELECTED AREAS IN COMMUNICATIONS, 2021, 39 (07) : 2150 - 2167
  • [4] Design and Analysis of Uplink and Downlink Communications for Federated Learning
    Zheng, Sihui
    Shen, Cong
    Chen, Xiang
    IEEE INTERNATIONAL CONFERENCE ON COMMUNICATIONS (ICC 2021), 2021,
  • [5] Enhanced Uplink Quantum Communication With Satellites via Downlink Channels
    Villaseñor E.
    He M.
    Wang Z.
    Malaney R.
    Win M.Z.
    IEEE Transactions on Quantum Engineering, 2021, 2
  • [6] Robust Federated Learning With Noisy Communication
    Ang, Fan
    Chen, Li
    Zhao, Nan
    Chen, Yunfei
    Wang, Weidong
    Yu, F. Richard
    IEEE TRANSACTIONS ON COMMUNICATIONS, 2020, 68 (06) : 3452 - 3464
  • [7] Robust Federated Learning Over Noisy Fading Channels
    Shah, Suhail M. M.
    Su, Liqun
    Lau, Vincent K. N.
    IEEE INTERNET OF THINGS JOURNAL, 2023, 10 (09) : 7993 - 8013
  • [8] Outage Probability Analysis for NOMA Downlink and Uplink Communication Systems With Generalized Fading Channels
    Agarwal, Akash
    Chaurasiya, Rishabh
    Rai, Sudhakar
    Jagannatham, Aditya K.
    IEEE ACCESS, 2020, 8 : 220461 - 220481
  • [9] Mix2FLD: Downlink Federated Learning After Uplink Federated Distillation With Two-Way Mixup
    Oh, Seungeun
    Park, Jihong
    Jeong, Eunjeong
    Kim, Hyesung
    Bennis, Mehdi
    Kim, Seong-Lyun
    IEEE COMMUNICATIONS LETTERS, 2020, 24 (10) : 2211 - 2215
  • [10] Decentralized Federated Learning Over Imperfect Communication Channels
    Li, Weicai
    Lv, Tiejun
    Ni, Wei
    Zhao, Jingbo
    Hossain, Ekram
    Poor, H. Vincent
    IEEE TRANSACTIONS ON COMMUNICATIONS, 2024, 72 (11) : 6973 - 6991