Factorized-FL: Agnostic Personalized Federated Learning with Kernel Factorization & Similarity Matching

TL;DR

Factorized-FL introduces a novel federated learning approach that handles highly heterogeneous client data by factorizing model parameters and selectively aggregating knowledge based on client similarity, improving personalization across diverse tasks.

Contribution

It proposes a new method that factorizes model parameters into shared and task-specific components, enabling effective personalization in label- and task-heterogeneous federated learning scenarios.

Findings

Outperforms state-of-the-art personalized FL methods on heterogeneous settings.

Effectively captures common and task-specific knowledge through parameter factorization.

Utilizes client similarity for selective aggregation, enhancing personalization.

Abstract

In real-world federated learning scenarios, participants could have their own personalized labels which are incompatible with those from other clients, due to using different label permutations or tackling completely different tasks or domains. However, most existing FL approaches cannot effectively tackle such extremely heterogeneous scenarios since they often assume that (1) all participants use a synchronized set of labels, and (2) they train on the same task from the same domain. In this work, to tackle these challenges, we introduce Factorized-FL, which allows to effectively tackle label- and task-heterogeneous federated learning settings by factorizing the model parameters into a pair of vectors, where one captures the common knowledge across different labels and tasks and the other captures knowledge specific to the task each local model tackles. Moreover, based on the distance in the client-specific vector space, Factorized-FL performs selective aggregation scheme to utilize only the knowledge from the relevant participants for each client. We extensively validate our method on both label- and domain-heterogeneous settings, on which it outperforms the state-of-the-art personalized federated learning methods.