Differentially Private Federated Learning ...
Document type :
Pré-publication ou Document de travail
Title :
Differentially Private Federated Learning on Heterogeneous Data
Author(s) :
Noble, Maxence [Auteur]
Centre de Mathématiques Appliquées de l'Ecole polytechnique [CMAP]
Bellet, Aurelien [Auteur]
Machine Learning in Information Networks [MAGNET]
Dieuleveut, Aymeric [Auteur]
Centre de Mathématiques Appliquées de l'Ecole polytechnique [CMAP]
Centre de Mathématiques Appliquées de l'Ecole polytechnique [CMAP]
Bellet, Aurelien [Auteur]

Machine Learning in Information Networks [MAGNET]
Dieuleveut, Aymeric [Auteur]
Centre de Mathématiques Appliquées de l'Ecole polytechnique [CMAP]
HAL domain(s) :
Informatique [cs]/Apprentissage [cs.LG]
Statistiques [stat]/Machine Learning [stat.ML]
Statistiques [stat]/Machine Learning [stat.ML]
English abstract : [en]
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. ...
Show more >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.Show less >
Show more >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.Show less >
Language :
Anglais
ANR Project :
Collections :
Source :
Files
- https://hal.inria.fr/hal-03498158/document
- Open access
- Access the document
- http://arxiv.org/pdf/2111.09278
- Open access
- Access the document
- https://hal.inria.fr/hal-03498158/document
- Open access
- Access the document
- https://hal.inria.fr/hal-03498158/document
- Open access
- Access the document
- document
- Open access
- Access the document
- 2111.09278.pdf
- Open access
- Access the document
- 2111.09278
- Open access
- Access the document
- document
- Open access
- Access the document
- 2111.09278.pdf
- Open access
- Access the document