Decentralized Personalized Federated Learning: Lower Bounds and Optimal Algorithm For All Personalization Modes
Archived with thanks to ScienceDirect
License: CC by NC-ND 4.0
Uploaded 03 November 2022
IR Deposit conditions:
OA version (pathway b) Accepted version
License: CC BY-NC-ND
No embargo
Published source must be acknowledged
Must link to publisher version with DOI
Abstract
This paper considers the problem of decentralized, personalized federated learning. For centralized personalized federated learning, a penalty that measures the deviation from the local model and its average, is often added to the objective function. However, in a decentralized setting this penalty is expensive in terms of communication costs, so here, a different penalty — one that is built to respect the structure of the underlying computational network — is used instead. We present lower bounds on the communication and local computation costs for this problem formulation and we also present provably optimal methods for decentralized personalized federated learning. Numerical experiments are presented to demonstrate the practical performance of our methods. © 2022 The Authors