Federated Learning with Erroneous Communication Links

01/31/2022

∙

In this paper, 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 e and 1-e, respectively. We provide mathematical proof for the convergence of the FL algorithm in the presence of communication errors, where the CN uses past local updates when the fresh updates are not received from some devices. We show via simulations that by using the past local updates, the FL algorithm can converge in the presence of communication errors. 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.

READ FULL TEXT

Federated Learning with Erroneous Communication Links

Federated Learning With Quantized Global Model Updates

To Talk or to Work: Delay Efficient Federated Learning over Mobile Edge Devices

Federated learning with hierarchical clustering of local updates to improve training on non-IID data

Communication Optimization in Large Scale Federated Learning using Autoencoder Compressed Weight Updates

Adaptive Gradient Sparsification for Efficient Federated Learning: An Online Learning Approach

FedShuffle: Recipes for Better Use of Local Work in Federated Learning

Adaptive Differential Filters for Fast and Communication-Efficient Federated Learning

Federated Learning with Erroneous Communication Links

Related Research

Federated Learning With Quantized Global Model Updates

To Talk or to Work: Delay Efficient Federated Learning over Mobile Edge Devices

Federated learning with hierarchical clustering of local updates to improve training on non-IID data

Communication Optimization in Large Scale Federated Learning using Autoencoder Compressed Weight Updates

Adaptive Gradient Sparsification for Efficient Federated Learning: An Online Learning Approach

FedShuffle: Recipes for Better Use of Local Work in Federated Learning

Adaptive Differential Filters for Fast and Communication-Efficient Federated Learning