FedSheafHN: Personalized Federated Learning on Graph-structured Data

TL;DR

FedSheafHN introduces a novel federated learning framework for graph data that leverages sheaf diffusion and collaboration graph embedding to personalize models effectively, outperforming existing methods in diverse scenarios.

Contribution

The paper proposes FedSheafHN, a new personalized federated learning model using sheaf diffusion and collaboration graph embedding for improved client-specific graph neural network customization.

Findings

Outperforms existing methods in client model accuracy

Achieves fast convergence across datasets

Generalizes well to new clients

Abstract

Personalized subgraph Federated Learning (FL) is a task that customizes Graph Neural Networks (GNNs) to individual client needs, accommodating diverse data distributions. However, applying hypernetworks in FL, while aiming to facilitate model personalization, often encounters challenges due to inadequate representation of client-specific characteristics. To overcome these limitations, we propose a model called FedSheafHN, using enhanced collaboration graph embedding and efficient personalized model parameter generation. Specifically, our model embeds each client's local subgraph into a server-constructed collaboration graph. We utilize sheaf diffusion in the collaboration graph to learn client representations. Our model improves the integration and interpretation of complex client characteristics. Furthermore, our model ensures the generation of personalized models through advanced hypernetworks optimized for parallel operations across clients. Empirical evaluations demonstrate that FedSheafHN outperforms existing methods in most scenarios, in terms of client model performance on various graph-structured datasets. It also has fast model convergence and effective new clients generalization.