GPFL: Simultaneously Learning Global and Personalized Feature Information for Personalized Federated Learning

TL;DR

GPFL introduces a novel personalized federated learning approach that simultaneously learns global and personalized features, improving effectiveness, scalability, fairness, stability, and privacy across diverse datasets.

Contribution

The paper proposes GPFL, a new method for personalized federated learning that extracts both global and personalized features simultaneously, addressing limitations of existing methods.

Findings

GPFL outperforms ten state-of-the-art methods in accuracy.

GPFL demonstrates superior scalability, fairness, and stability.

GPFL reduces overfitting and enhances privacy.

Abstract

Federated Learning (FL) is popular for its privacy-preserving and collaborative learning capabilities. Recently, personalized FL (pFL) has received attention for its ability to address statistical heterogeneity and achieve personalization in FL. However, from the perspective of feature extraction, most existing pFL methods only focus on extracting global or personalized feature information during local training, which fails to meet the collaborative learning and personalization goals of pFL. To address this, we propose a new pFL method, named GPFL, to simultaneously learn global and personalized feature information on each client. We conduct extensive experiments on six datasets in three statistically heterogeneous settings and show the superiority of GPFL over ten state-of-the-art methods regarding effectiveness, scalability, fairness, stability, and privacy. Besides, GPFL mitigates overfitting and outperforms the baselines by up to 8.99% in accuracy.