Differentially Private Federated Learning on Heterogeneous Data - Archive ouverte HAL Access content directly
Preprints, Working Papers, ... Year :

Differentially Private Federated Learning on Heterogeneous Data

(1) , (2) , (1)
1
2

Abstract

Federated Learning (FL) is a paradigm for large-scale distributed learning which faces two key challenges: (i) efficient training from highly heterogeneous user data, and (ii) protecting the privacy of participating users. In this work, we propose a novel FL approach (DP-SCAFFOLD) to tackle these two challenges together by incorporating Differential Privacy (DP) constraints into the popular SCAFFOLD algorithm. We focus on the challenging setting where users communicate with a ''honest-but-curious'' server without any trusted intermediary, which requires to ensure privacy not only towards a third-party with access to the final model but also towards the server who observes all user communications. Using advanced results from DP theory, we establish the convergence of our algorithm for convex and non-convex objectives. Our analysis clearly highlights the privacy-utility trade-off under data heterogeneity, and demonstrates the superiority of DP-SCAFFOLD over the state-of-the-art algorithm DP-FedAvg when the number of local updates and the level of heterogeneity grow. Our numerical results confirm our analysis and show that DP-SCAFFOLD provides significant gains in practice.
Fichier principal
Vignette du fichier
2111.09278.pdf (1.88 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03498158 , version 1 (20-12-2021)

Identifiers

Cite

Maxence Noble, Aurélien Bellet, Aymeric Dieuleveut. Differentially Private Federated Learning on Heterogeneous Data. 2021. ⟨hal-03498158⟩
73 View
109 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More