Curriculum Guided Personalized Subgraph Federated Learning

2025-08-30

AI Summary

Subgraph federated learning (Subgraph FL) suffers from local overfitting and distorted structural similarity estimation due to data heterogeneity, leading to ineffective weighted aggregation and hindered knowledge fusion. To address these challenges, we propose CUFL—a curriculum-guided personalized federated learning framework. First, we design an edge-level curriculum learning strategy based on reconstruction scores, enabling progressive training from general graph structures to client-specific patterns. Second, we introduce a fine-grained structural similarity metric computed over randomly sampled reference graphs, enhancing the robustness of client weight estimation. Third, we implement dynamic weighted aggregation to facilitate cross-client knowledge transfer. Evaluated on six benchmark datasets, CUFL consistently outperforms state-of-the-art methods, achieving simultaneous improvements in generalization and personalization performance. The source code is publicly available.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Search and Optimization: Learning to Search
• Computer Vision: Learning & Optimization for CV

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Subgraph Federated Learning (FL) aims to train Graph Neural Networks (GNNs) across distributed private subgraphs, but it suffers from severe data heterogeneity. To mitigate data heterogeneity, weighted model aggregation personalizes each local GNN by assigning larger weights to parameters from clients with similar subgraph characteristics inferred from their current model states. However, the sparse and biased subgraphs often trigger rapid overfitting, causing the estimated client similarity matrix to stagnate or even collapse. As a result, aggregation loses effectiveness as clients reinforce their own biases instead of exploiting diverse knowledge otherwise available. To this end, we propose a novel personalized subgraph FL framework called Curriculum guided personalized sUbgraph Federated Learning (CUFL). On the client side, CUFL adopts Curriculum Learning (CL) that adaptively selects edges for training according to their reconstruction scores, exposing each GNN first to easier, generic cross-client substructures and only later to harder, client-specific ones. This paced exposure prevents early overfitting to biased patterns and enables gradual personalization. By regulating personalization, the curriculum also reshapes server aggregation from exchanging generic knowledge to propagating client-specific knowledge. Further, CUFL improves weighted aggregation by estimating client similarity using fine-grained structural indicators reconstructed on a random reference graph. Extensive experiments on six benchmark datasets confirm that CUFL achieves superior performance compared to relevant baselines. Code is available at https://github.com/Kang-Min-Ku/CUFL.git.

Problem

• Addresses data heterogeneity in subgraph federated learning
• Mitigates rapid overfitting in distributed GNN training
• Improves client similarity estimation for model aggregation

Innovation

• Curriculum Learning adaptively selects edges by reconstruction scores
• Fine-grained structural indicators estimate client similarity accurately
• Paced exposure prevents overfitting and enables gradual personalization