Federated Learning With Erroneous Communication Links
Federated Learning With Erroneous Communication Links
复制标题
通信链路错误的联邦学习
DOI：
--
复制
发表时间：
2022
期刊：
影响因子：
--
通讯作者：
Aradhika Guha
中科院分区：
文献类型：
--
作者：
M. Shirvanimoghaddam;Ayoob Salari;Yifeng Gao;Aradhika Guha
在这封信中，我们考虑了存在通信错误的联邦学习（FL）问题。我们建模的设备和中心节点（CN）之间的链路由一个数据包擦除通道，其中本地参数从设备被删除或正确接收CN的概率分别为<inline-formula><tex-math notation="LaTeX">$100 $</tex-math></inline-formula>和<inline-formula><tex-math notation="LaTeX">$1- 100 $</tex-math></inline-formula>。我们证明了FL算法在存在通信错误的情况下，CN使用过去的本地更新，如果没有从设备接收到新的本地更新，则FL算法收敛到相同的全局参数，而没有任何通信错误。我们提供了几个仿真结果来验证我们的理论分析。我们还表明，当数据集均匀分布在设备之间时，仅使用新鲜更新并丢弃丢失更新的FL算法可能比使用过去本地更新的FL算法收敛得更快。
In this letter, we consider the federated learning (FL) problem in the presence of communication errors. We model the link between the devices and the central node (CN) by a packet erasure channel, where the local parameters from devices are either erased or received correctly by CN with probability <inline-formula> <tex-math notation="LaTeX">$epsilon $ </tex-math></inline-formula> and <inline-formula> <tex-math notation="LaTeX">$1-epsilon $ </tex-math></inline-formula>, respectively. We proved that the FL algorithm in the presence of communication errors, where the CN uses the past local update if the fresh one is not received from a device, converges to the same global parameter as that the FL algorithm converges to without any communication error. We provide several simulation results to validate our theoretical analysis. We also show that when the dataset is uniformly distributed among devices, the FL algorithm that only uses fresh updates and discards missing updates might converge faster than the FL algorithm that uses past local updates.
影响因子：
10.4
作者：
Amiri, Mohammad Mohammadi;Gunduz, Deniz;Poor, H. Vincent
通讯作者：
Poor, H. Vincent