FairGFL: Privacy-Preserving Fairness-Aware Federated Learning with Overlapping Subgraphs
Zihao Zhou, Shusen Yang, Fangyuan Zhao, Xuebin Ren
TL;DR
This paper tackles fairness in graph federated learning when clients hold overlapping subgraphs that are imbalanced. It introduces FairGFL, a privacy-preserving framework that privately estimates overlapping ratios via Local Differential Privacy, then uses an interpretable fairness-aware aggregation and a composite loss with a max-loss regularizer to balance fairness and utility. The approach yields improved cross-client fairness (lower variance, higher entropy) while preserving model utility across four benchmark graph datasets, under various privacy budgets and overlap scenarios. The work advances practical, privacy-preserving collaborative graph learning by addressing overlap-induced heterogeneity and providing convergence analysis and empirical validation. The methods have potential impact for real-world federated graph applications requiring fair, privacy-conscious collaboration.
Abstract
Graph federated learning enables the collaborative extraction of high-order information from distributed subgraphs while preserving the privacy of raw data. However, graph data often exhibits overlap among different clients. Previous research has demonstrated certain benefits of overlapping data in mitigating data heterogeneity. However, the negative effects have not been explored, particularly in cases where the overlaps are imbalanced across clients. In this paper, we uncover the unfairness issue arising from imbalanced overlapping subgraphs through both empirical observations and theoretical reasoning. To address this issue, we propose FairGFL (FAIRness-aware subGraph Federated Learning), a novel algorithm that enhances cross-client fairness while maintaining model utility in a privacy-preserving manner. Specifically, FairGFL incorporates an interpretable weighted aggregation approach to enhance fairness across clients, leveraging privacy-preserving estimation of their overlapping ratios. Furthermore, FairGFL improves the tradeoff between model utility and fairness by integrating a carefully crafted regularizer into the federated composite loss function. Through extensive experiments on four benchmark graph datasets, we demonstrate that FairGFL outperforms four representative baseline algorithms in terms of both model utility and fairness.