S2FGL: Spatial Spectral Federated Graph Learning

2025-07-03

AI Summary

This work addresses two critical challenges in Federated Graph Learning (FGL): (1) label signal disruption due to structural disconnection across local subgraphs, and (2) spectral heterogeneity inducing client-wise frequency-domain drift. To this end, we propose the first dual-perspective (spatial and spectral) framework for FGL. Spatially, we introduce a global knowledge base that caches and propagates class-level structural patterns shared across clients, mitigating structural disconnection. Spectrally, we design a cross-client frequency alignment mechanism to calibrate the spectral biases of local GNNs and suppress spectral drift. Our approach integrates spectral graph theory, federated optimization, and dynamic knowledge updating. Extensive experiments on multiple benchmark datasets demonstrate significant improvements in both model accuracy and convergence stability, with superior generalization over existing state-of-the-art methods. The source code is publicly available.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Federated Graph Learning (FGL) combines the privacy-preserving capabilities of federated learning (FL) with the strong graph modeling capability of Graph Neural Networks (GNNs). Current research addresses subgraph-FL only from the structural perspective, neglecting the propagation of graph signals on spatial and spectral domains of the structure. From a spatial perspective, subgraph-FL introduces edge disconnections between clients, leading to disruptions in label signals and a degradation in the class knowledge of the global GNN. From a spectral perspective, spectral heterogeneity causes inconsistencies in signal frequencies across subgraphs, which makes local GNNs overfit the local signal propagation schemes. As a result, spectral client drifts occur, undermining global generalizability. To tackle the challenges, we propose a global knowledge repository to mitigate label signal disruption and a frequency alignment to address spectral client drifts. The combination of spatial and spectral strategies forms our framework S2FGL. Extensive experiments on multiple datasets demonstrate the superiority of S2FGL. The code is available at https://github.com/Wonder7racer/S2FGL.git.

Problem

• Addresses edge disconnections disrupting label signals in subgraph-FL
• Mitigates spectral heterogeneity causing local GNN overfitting
• Combats spectral client drifts undermining global model generalizability

Innovation

• Global knowledge repository mitigates label disruption
• Frequency alignment addresses spectral client drifts
• Combines spatial and spectral strategies in S2FGL