SAFA: a Semi-Asynchronous Protocol for Fast Federated Learning with Low Overhead

TL;DR

SAFA is a semi-asynchronous federated learning protocol designed to enhance efficiency, robustness, and model accuracy in unreliable device environments while maintaining low communication overhead.

Contribution

The paper introduces a novel semi-asynchronous protocol with new strategies for model distribution, client selection, and aggregation to address stragglers and staleness in federated learning.

Findings

Reduces federated round duration

Improves global model accuracy

Lowers local resource wastage

Abstract

Federated learning (FL) has attracted increasing attention as a promising approach to driving a vast number of end devices with artificial intelligence. However, it is very challenging to guarantee the efficiency of FL considering the unreliable nature of end devices while the cost of device-server communication cannot be neglected. In this paper, we propose SAFA, a semi-asynchronous FL protocol, to address the problems in federated learning such as low round efficiency and poor convergence rate in extreme conditions (e.g., clients dropping offline frequently). We introduce novel designs in the steps of model distribution, client selection and global aggregation to mitigate the impacts of stragglers, crashes and model staleness in order to boost efficiency and improve the quality of the global model. We have conducted extensive experiments with typical machine learning tasks. The results demonstrate that the proposed protocol is effective in terms of shortening federated round duration, reducing local resource wastage, and improving the accuracy of the global model at an acceptable communication cost.