FedGKD: Unleashing the Power of Collaboration in Federated Graph Neural Networks

TL;DR

FedGKD introduces a new federated GNN framework that enhances collaboration by better quantifying task relatedness and exploiting global collaboration structures, leading to improved performance across diverse datasets.

Contribution

The paper presents FedGKD, a novel federated GNN framework with client-side graph dataset distillation and server-side structure-aware aggregation, addressing heterogeneity challenges.

Findings

Outperforms existing methods on six real-world datasets

Improves task relatedness quantification

Enhances collaboration efficiency in federated GNNs

Abstract

Federated training of Graph Neural Networks (GNN) has become popular in recent years due to its ability to perform graph-related tasks under data isolation scenarios while preserving data privacy. However, graph heterogeneity issues in federated GNN systems continue to pose challenges. Existing frameworks address the problem by representing local tasks using different statistics and relating them through a simple aggregation mechanism. However, these approaches suffer from limited efficiency from two aspects: low quality of task-relatedness quantification and inefficacy of exploiting the collaboration structure. To address these issues, we propose FedGKD, a novel federated GNN framework that utilizes a novel client-side graph dataset distillation method to extract task features that better describe task-relatedness, and introduces a novel server-side aggregation mechanism that is aware of the global collaboration structure. We conduct extensive experiments on six real-world datasets of different scales, demonstrating our framework's outperformance.