FedCIA: Federated Collaborative Information Aggregation for Privacy-Preserving Recommendation

2025-04-19

AI Summary

Existing federated recommendation algorithms employ uniform weighted aggregation of client-side item embeddings, leading to information loss, weakened personalization, and incompatibility with parameter-free models. Method: We propose a novel collaborative information aggregation paradigm that abandons parameter-level aggregation; instead, clients upload item similarity matrices, enabling cross-device collaborative signal sharing—rather than embedding-space alignment—while preserving local personalized distributions. Contribution/Results: This approach is inherently compatible with parameter-free recommendation models and enjoys theoretically guaranteed information fidelity. Extensive experiments on multiple real-world datasets demonstrate that our method significantly outperforms state-of-the-art federated recommendation approaches, achieving superior trade-offs among recommendation accuracy, personalization preservation, and privacy protection.

Technology Category

• Machine Learning: Distributed Machine Learning & Federated Learning
• Multiagent Systems: Mechanism Design
• Data Mining & Knowledge Management: Recommender Systems

Application Category

• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Search and Retrieval-Augmented AI: Personalized, context-aware and across-device search

Abstract

Recommendation algorithms rely on user historical interactions to deliver personalized suggestions, which raises significant privacy concerns. Federated recommendation algorithms tackle this issue by combining local model training with server-side model aggregation, where most existing algorithms use a uniform weighted summation to aggregate item embeddings from different client models. This approach has three major limitations: 1) information loss during aggregation, 2) failure to retain personalized local features, and 3) incompatibility with parameter-free recommendation algorithms. To address these limitations, we first review the development of recommendation algorithms and recognize that their core function is to share collaborative information, specifically the global relationship between users and items. With this understanding, we propose a novel aggregation paradigm named collaborative information aggregation, which focuses on sharing collaborative information rather than item parameters. Based on this new paradigm, we introduce the federated collaborative information aggregation (FedCIA) method for privacy-preserving recommendation. This method requires each client to upload item similarity matrices for aggregation, which allows clients to align their local models without constraining embeddings to a unified vector space. As a result, it mitigates information loss caused by direct summation, preserves the personalized embedding distributions of individual clients, and supports the aggregation of parameter-free models. Theoretical analysis and experimental results on real-world datasets demonstrate the superior performance of FedCIA compared with the state-of-the-art federated recommendation algorithms. Code is available at https://github.com/Mingzhe-Han/FedCIA.

Problem

• Addresses privacy concerns in recommendation algorithms
• Improves aggregation of item embeddings in federated learning
• Enhances personalized features and supports parameter-free models

Innovation

• Federated collaborative information aggregation method
• Uploads item similarity matrices for aggregation
• Preserves personalized local embedding distributions