Low-pass Personalized Subgraph Federated Recommendation

2026-03-20

AI Summary

This work addresses the challenge in federated recommendation systems where significant heterogeneity in client subgraphs—both in size and connectivity—hinders representation alignment, thereby compromising model robustness and personalization. To mitigate this, the paper introduces, for the first time, graph Fourier transform combined with low-pass spectral filtering to extract stable low-frequency structural signals across clients. It further designs neutral structural anchors to guide personalized parameter updates and proposes a local popularity bias-aware boundary correction term to alleviate data-induced biases. The proposed method effectively handles structural heterogeneity among client subgraphs and consistently outperforms existing approaches across five real-world datasets, achieving a superior balance among recommendation accuracy, fairness, and robustness.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Local Search

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

Federated Recommender Systems (FRS) preserve privacy by training decentralized models on client-specific user-item subgraphs without sharing raw data. However, FRS faces a unique challenge: subgraph structural imbalance, where drastic variations in subgraph scale (user/item counts) and connectivity (item degree) misalign client representations, making it challenging to train a robust model that respects each client's unique structural characteristics. To address this, we propose a Low-pass Personalized Subgraph Federated recommender system (LPSFed). LPSFed leverages graph Fourier transforms and low-pass spectral filtering to extract low-frequency structural signals that remain stable across subgraphs of varying size and degree, allowing robust personalized parameter updates guided by similarity to a neutral structural anchor. Additionally, we leverage a localized popularity bias-aware margin that captures item-degree imbalance within each subgraph and incorporates it into a personalized bias correction term to mitigate recommendation bias. Supported by theoretical analysis and validated on five real-world datasets, LPSFed achieves superior recommendation accuracy and enhances model robustness.

Problem

• federated recommender systems
• subgraph structural imbalance
• personalized recommendation
• graph heterogeneity
• privacy-preserving recommendation

Innovation

• federated recommendation
• subgraph structural imbalance
• graph Fourier transform
• low-pass spectral filtering
• popularity bias correction