Personalized Federated Learning with First Order Model Optimization

TL;DR

This paper introduces a personalized federated learning approach that optimizes client-specific model combinations without assuming data distribution knowledge, leading to improved performance and flexibility over traditional methods.

Contribution

It proposes a novel method for personalized federated learning that computes optimal weighted model combinations for each client, enhancing personalization without requiring data distribution assumptions.

Findings

Outperforms existing personalized FL methods across various datasets.

Enables transfer learning outside local data distributions.

Supports arbitrary target distributions for personalization.

Abstract

While federated learning traditionally aims to train a single global model across decentralized local datasets, one model may not always be ideal for all participating clients. Here we propose an alternative, where each client only federates with other relevant clients to obtain a stronger model per client-specific objectives. To achieve this personalization, rather than computing a single model average with constant weights for the entire federation as in traditional FL, we efficiently calculate optimal weighted model combinations for each client, based on figuring out how much a client can benefit from another's model. We do not assume knowledge of any underlying data distributions or client similarities, and allow each client to optimize for arbitrary target distributions of interest, enabling greater flexibility for personalization. We evaluate and characterize our method on a variety of federated settings, datasets, and degrees of local data heterogeneity. Our method outperforms existing alternatives, while also enabling new features for personalized FL such as transfer outside of local data distributions.