English

Fast Server Learning Rate Tuning for Coded Federated Dropout

Machine Learning 2022-09-16 v4

Abstract

In cross-device Federated Learning (FL), clients with low computational power train a common\linebreak[4] machine model by exchanging parameters via updates instead of potentially private data. Federated Dropout (FD) is a technique that improves the communication efficiency of a FL session by selecting a \emph{subset} of model parameters to be updated in each training round. However, compared to standard FL, FD produces considerably lower accuracy and faces a longer convergence time. In this paper, we leverage \textit{coding theory} to enhance FD by allowing different sub-models to be used at each client. We also show that by carefully tuning the server learning rate hyper-parameter, we can achieve higher training speed while also achieving up to the same final accuracy as the no dropout case. For the EMNIST dataset, our mechanism achieves 99.6\% of the final accuracy of the no dropout case while requiring 2.43×2.43\times less bandwidth to achieve this level of accuracy.

Keywords

Cite

@article{arxiv.2201.11036,
  title  = {Fast Server Learning Rate Tuning for Coded Federated Dropout},
  author = {Giacomo Verardo and Daniel Barreira and Marco Chiesa and Dejan Kostic and Gerald Q. Maguire},
  journal= {arXiv preprint arXiv:2201.11036},
  year   = {2022}
}

Comments

6 pages plus references and appendix, 6 figures. Accepted and presented at FL-IJCAI22 (https://federated-learning.org/fl-ijcai-2022/)

R2 v1 2026-06-24T09:03:58.824Z