Federated Learning with Partial Model Personalization

Michael Rabbat, Abdelrahman Mohamed, Kshitiz Malik, Maziar Sanjabi, Krishna Pillutla, Lin Xiao

Publication date: 7 April 2022

Abstract: We consider two federated learning algorithms for training partially personalized models, where the shared and personal parameters are updated either simultaneously or alternately on the devices. Both algorithms have been proposed in the literature, but their convergence properties are not fully understood, especially for the alternating variant. We provide convergence analyses of both algorithms in the general nonconvex setting with partial participation and delineate the regime where one dominates the other. Our experiments on real-world image, text, and speech datasets demonstrate that (a) partial personalization can obtain most of the benefits of full model personalization with a small fraction of personal parameters, and, (b) the alternating update algorithm often outperforms the simultaneous update algorithm by a small but consistent margin.

Has companion code repository: https://github.com/facebookresearch/fl_partial_personalization

This page was built for publication: Federated Learning with Partial Model Personalization