Optimizing Federated Averaging over Fading Channels
Optimizing Federated Averaging over Fading Channels
复制标题
DOI：
10.1109/isit50566.2022.9834609
复制
发表时间：
2022-06
期刊：
影响因子：
--
通讯作者：
Yujia Mu;Cong Shen;Yonina C. Eldar
中科院分区：
文献类型：
--
作者：
Yujia Mu;Cong Shen;Yonina C. Eldar
深度衰落表示在无线信道上通信时的典型错误事件。我们表明，深度衰落是特别有害的联邦学习（FL）通过无线通信。特别是，著名的FEDVG和它的几个变种打破FL任务时，深衰落存在于通信阶段。本文的主要贡献是在参数服务器上的最优全局模型聚合方法，该方法不仅根据客户端的学习特性，而且根据接收端的瞬时信道状态信息（CSIR）为不同的客户端分配不同的权重。这是通过首先推导衰落信道上的并行随机梯度下降（SGD）收敛的上界，然后求解最小化该上界的服务器聚合权重的优化问题来实现的。导出的最优聚合解是封闭形式的，并实现了众所周知的O（1/t）收敛速度强凸损失函数在任意衰落和衰减的学习率。我们使用几个现实世界的FL任务验证我们的方法。
Deep fading represents the typical error event when communicating over wireless channels. We show that deep fading is particularly detrimental for federated learning (FL) over wireless communications. In particular, the celebrated FEDAVG and several of its variants break down for FL tasks when deep fading exists in the communication phase. The main contribution of this paper is an optimal global model aggregation method at the parameter server, which allocates different weights to different clients based on not only their learning characteristics but also the instantaneous channel state information at the receiver (CSIR). This is accomplished by first deriving an upper bound on the parallel stochastic gradient descent (SGD) convergence over fading channels, and then solving an optimization problem for the server aggregation weights that minimizes this upper bound. The derived optimal aggregation solution is closed-form, and achieves the well-known O(1/t) convergence rate for strongly-convex loss functions under arbitrary fading and decaying learning rates. We validate our approach using several real-world FL tasks.