Vertically Federated Graph Neural Network for Privacy-Preserving Node Classification

TL;DR

This paper introduces VFGNN, a federated GNN framework that enables privacy-preserving node classification on vertically partitioned graph data, combining differential privacy and distributed computation.

Contribution

It proposes a novel federated GNN paradigm for privacy preservation in vertically partitioned graphs, adaptable to existing GNN models.

Findings

VFGNN effectively preserves privacy during node classification.

Experimental results show competitive accuracy on benchmark datasets.

Differential privacy enhances security against data leakage.

Abstract

Recently, Graph Neural Network (GNN) has achieved remarkable progresses in various real-world tasks on graph data, consisting of node features and the adjacent information between different nodes. High-performance GNN models always depend on both rich features and complete edge information in graph. However, such information could possibly be isolated by different data holders in practice, which is the so-called data isolation problem. To solve this problem, in this paper, we propose VFGNN, a federated GNN learning paradigm for privacy-preserving node classification task under data vertically partitioned setting, which can be generalized to existing GNN models. Specifically, we split the computation graph into two parts. We leave the private data (i.e., features, edges, and labels) related computations on data holders, and delegate the rest of computations to a semi-honest server. We also propose to apply differential privacy to prevent potential information leakage from the server. We conduct experiments on three benchmarks and the results demonstrate the effectiveness of VFGNN.