Lossless and Privacy-Preserving Graph Convolution Network for Federated Item Recommendation

TL;DR

This paper introduces LP-GCN, a novel federated graph neural network that fully preserves privacy while achieving recommendation performance comparable to centralized methods, validated through theoretical and empirical analysis.

Contribution

It proposes a lossless, privacy-preserving GCN for federated recommendation that fully completes the graph convolution process without privacy leakage.

Findings

LP-GCN matches centralized GCN performance.

Outperforms existing federated recommendation methods.

Validated through extensive experiments on real-world datasets.

Abstract

Graph neural network (GNN) has emerged as a state-of-the-art solution for item recommendation. However, existing GNN-based recommendation methods rely on a centralized storage of fragmented user-item interaction sub-graphs and training on an aggregated global graph, which will lead to privacy concerns. As a response, some recent works develop GNN-based federated recommendation methods by exploiting decentralized and fragmented user-item sub-graphs in order to preserve user privacy. However, due to privacy constraints, the graph convolution process in existing federated recommendation methods is incomplete compared with the centralized counterpart, causing a degradation of the recommendation performance. In this paper, we propose a novel lossless and privacy-preserving graph convolution network (LP-GCN), which fully completes the graph convolution process with decentralized user-item interaction sub-graphs while ensuring privacy. It is worth mentioning that its performance is equivalent to that of the non-federated (i.e., centralized) counterpart. Moreover, we validate its effectiveness through both theoretical analysis and empirical studies. Extensive experiments on three real-world datasets show that our LP-GCN outperforms the existing federated recommendation methods. The code will be publicly available once the paper is accepted.