SAFA: a semi-asynchronous protocol for fast federated learning with low overhead

Research output: Contribution to journalArticlepeer-review

Authors

  • Wentai Wu
  • Ligang He
  • Weiwei Lin
  • Rui Mao
  • Carsten Maple

Colleges, School and Institutes

External organisations

  • University of Warwick
  • South China University of Technology
  • Shenzhen University, Shenzhen, China.

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 article, 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.

Bibliographic note

Funding Information: This work was supported in part by Worldwide Byte Security Information Technology Company ltd., in part by National Natural Science Foundation of China under Grant 61772205, in part by Guangzhou Development Zone Science and Technology under Grant 2018GH17, in part by Major Program and of Guangdong Basic and Applied Research under Grant 2019B030302002, in part by Guangdong project under Grant 2017B030314073 and Grant 2018B030325002, in part by the EPSRC Centre for Doctoral Training in Urban Science under EPSRC Grant EP/L016400/1, in part by the Alan Turing Institute under EPSRC Grant EP/N510129/1 and Grant PETRAS, and in part by the National Center of Excellence for IoT Systems Cybersecurity under Grant EP/ S035362/1. Publisher Copyright: © 1968-2012 IEEE.

Details

Original languageEnglish
Article number9093123
Pages (from-to)655-668
Number of pages14
JournalIEEE Transactions on Computers
Volume70
Issue number5
Early online date14 May 2020
Publication statusPublished - 1 May 2021

Keywords

  • Distributed computing, edge intelligence, federated learning, machine learning