Federated Hypergraph Learning: Hyperedge Completion with Local Differential Privacy

2024-08-09

AI Summary

In distributed storage settings, hypergraph fragmentation impedes high-order relational modeling and compromises local information integrity. Method: We propose FedHGL, a federated hypergraph learning framework that jointly addresses hyperedge completion and high-order relation modeling under data silos. Its core innovations are: (1) a novel pre-propagation hyperedge completion mechanism that enables privacy-preserving, cross-client aggregation of high-order features prior to federated training; and (2) integration of local differential privacy (LDP) to prevent leakage of raw node features. FedHGL supports collaborative training of hypergraph neural networks across multiple clients without sharing raw data. Results: Extensive experiments on seven real-world datasets demonstrate that FedHGL significantly outperforms existing federated graph learning baselines, effectively mitigating high-order information loss in distributed scenarios, while improving both hyperedge completion accuracy and downstream task performance.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Data Mining & Knowledge Management: Linked Open Data, Knowledge Graphs & KB Completion
• 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
• Systems and Infrastructure for Web, Mobile and WoT: Federated Web and WoT systems, including distributed, federated and edge-based data processing

Abstract

As the volume and complexity increase, graph-structured data commonly need to be split and stored across distributed systems. To enable data mining on subgraphs within these distributed systems, federated graph learning has been proposed, allowing collaborative training of Graph Neural Networks (GNNs) across clients without sharing raw node features. However, when dealing with graph structures that involve high-order relationships between nodes, known as hypergraphs, existing federated graph learning methods are less effective. In this study, we introduce FedHGL, an innovative federated hypergraph learning algorithm. FedHGL is designed to collaboratively train a comprehensive hypergraph neural network across multiple clients, facilitating mining tasks on subgraphs of a hypergraph where relationships are not merely pairwise. To address the high-order information loss between subgraphs caused by distributed storage, we introduce a pre-propagation hyperedge completion operation before the federated training process. In this pre-propagation step, cross-client feature aggregation is performed and distributed at the central server to ensure that this information can be utilized by the clients. Furthermore, by incorporating local differential privacy (LDP) mechanisms, we ensure that the original node features are not disclosed during this aggregation process. Experimental results on seven real-world datasets confirm the effectiveness of our approach and demonstrate its performance advantages over traditional federated graph learning methods.

Problem

• Federated hypergraph learning with local privacy
• Preserving high-order structural integrity in partitions
• Protecting sensitive node features from inference attacks

Innovation

• Federated hypergraph learning with privacy guarantees
• Pre-propagation hyperedge completion for structural integrity
• Local differential privacy for node feature protection