Federated Learning under Heterogeneous and Correlated Client Availability - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Conference Papers Year : 2023

Federated Learning under Heterogeneous and Correlated Client Availability

Abstract

The enormous amount of data produced by mobile and IoT devices has motivated the development of federated learning (FL), a framework allowing such devices (or clients) to collaboratively train machine learning models without sharing their local data. FL algorithms (like FedAvg) iteratively aggregate model updates computed by clients on their own datasets. Clients may exhibit different levels of participation, often correlated over time and with other clients. This paper presents the first convergence analysis for a FedAvg-like FL algorithm under heterogeneous and correlated client availability. Our analysis highlights how correlation adversely affects the algorithm's convergence rate and how the aggregation strategy can alleviate this effect at the cost of steering training toward a biased model. Guided by the theoretical analysis, we propose CA-Fed, a new FL algorithm that tries to balance the conflicting goals of maximizing convergence speed and minimizing model bias. To this purpose, CA-Fed dynamically adapts the weight given to each client and may ignore clients with low availability and large correlation. Our experimental results show that CA-Fed achieves higher timeaverage accuracy and a lower standard deviation than state-ofthe-art AdaFed and F3AST, both on synthetic and real datasets.
Fichier principal
Vignette du fichier
infocom.pdf (577.82 Ko) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04364285 , version 1 (26-12-2023)

Licence

Identifiers

Cite

Angelo Rodio, Francescomaria Faticanti, Othmane Marfoq, Giovanni Neglia, Emilio Leonardi. Federated Learning under Heterogeneous and Correlated Client Availability. IEEE INFOCOM 2023 - IEEE Conference on Computer Communications, May 2023, New York City, United States. pp.1-10, ⟨10.1109/INFOCOM53939.2023.10228876⟩. ⟨hal-04364285⟩
65 View
18 Download

Altmetric

Share

Gmail Mastodon Facebook X LinkedIn More