Far From Sight, Far From Mind: Inverse Distance Weighting for Graph Federated Recommendation

TL;DR

This paper introduces Dist-FedAvg, a novel distance-based aggregation method for graph federated recommendation systems that improves personalization and recommendation accuracy by considering user similarity and anchor user influence.

Contribution

The paper proposes Dist-FedAvg, a new aggregation technique that accounts for user similarity and anchor influence, addressing limitations of traditional methods in graph federated recommendation.

Findings

Dist-FedAvg outperforms baseline methods in recommendation accuracy.

The method effectively balances personalization with anchor user influence.

Empirical results show consistent improvements across multiple datasets.

Abstract

Graph federated recommendation systems offer a privacy-preserving alternative to traditional centralized recommendation architectures, which often raise concerns about data security. While federated learning enables personalized recommendations without exposing raw user data, existing aggregation methods overlook the unique properties of user embeddings in this setting. Indeed, traditional aggregation methods fail to account for their complexity and the critical role of user similarity in recommendation effectiveness. Moreover, evolving user interactions require adaptive aggregation while preserving the influence of high-relevance anchor users (the primary users before expansion in graph-based frameworks). To address these limitations, we introduce Dist-FedAvg, a novel distance-based aggregation method designed to enhance personalization and aggregation efficiency in graph federated learning. Our method assigns higher aggregation weights to users with similar embeddings, while ensuring that anchor users retain significant influence in local updates. Empirical evaluations on multiple datasets demonstrate that Dist-FedAvg consistently outperforms baseline aggregation techniques, improving recommendation accuracy while maintaining seamless integration into existing federated learning frameworks.