Federated Learning over Noisy Channels
Federated Learning over Noisy Channels
复制标题
DOI：
10.1109/icc42927.2021.9500833
复制
发表时间：
2021-01
期刊：
影响因子：
--
通讯作者：
Xizixiang Wei;Cong Shen
中科院分区：
文献类型：
--
作者：
Xizixiang Wei;Cong Shen
当上行链路和下行链路通信都有错误时，联邦学习（FL）是否工作？英语能承受多大的交流噪音？它对学习成绩的影响是什么？这项工作致力于通过明确地在FL管道中合并上行链路和下行链路噪声通道来回答这些实际重要的问题。我们对同时上行和下行噪声通信信道上的FL进行了严格的收敛分析，并描述了FL保持相同收敛速率缩放的充分条件，作为无通信错误的理想情况。分析表明，为了保持fedag在完美通信条件下的$\mathcal{O}\left({1/T} \right)$收敛速率，应控制上行链路和下行链路的信噪比（SNR），使其按$\mathcal{O}\left({{T ^2}} \right)$的比例进行伸缩，其中T为通信轮数指标。这一关键结果导致模拟聚合的发射功率控制策略，其性能优于标准方法，通过使用真实世界的FL任务进行广泛的数值实验。
Does Federated Learning (FL) work when both uplink and downlink communications have errors? How much communication noise can FL handle and what is its impact to the learning performance? This work is devoted to answering these practically important questions by explicitly incorporating both uplink and downlink noisy channels in the FL pipeline. We present a rigorous convergence analysis of FL over simultaneous uplink and downlink noisy communication channels, and characterize the sufficient conditions for FL to maintain the same convergence rate scaling as the ideal case of no communication error. The analysis reveals that, in order to maintain the $\mathcal{O}\left( {1/T} \right)$ convergence rate of FedAvg with perfect communications, the uplink and downlink signal-to-noise-ratio (SNR) should be controlled such that they scale as $\mathcal{O}\left( {{t^2}} \right)$ where t is the index of communication rounds. This key result leads to a transmit power control policy for analog aggregation, whose performance is shown to be superior over the standard method via extensive numerical experiments using real-world FL tasks.