Towards Representation Identical Privacy-Preserving Graph Neural Network via Split Learning
This addresses privacy concerns for organizations using GNNs on sensitive distributed data, though it is an incremental extension of split learning to graph structures.
The paper tackles the challenge of privacy-preserving training and inference for Graph Neural Networks (GNNs) on distributed graph data, proposing SAPGNN, a split learning-based method that achieves the same accuracy as centralized GNNs while keeping data local.
In recent years, the fast rise in number of studies on graph neural network (GNN) has put it from the theories research to reality application stage. Despite the encouraging performance achieved by GNN, less attention has been paid to the privacy-preserving training and inference over distributed graph data in the related literature. Due to the particularity of graph structure, it is challenging to extend the existing private learning framework to GNN. Motivated by the idea of split learning, we propose a \textbf{S}erver \textbf{A}ided \textbf{P}rivacy-preserving \textbf{GNN} (SAPGNN) for the node level task on horizontally partitioned cross-silo scenario. It offers a natural extension of centralized GNN to isolated graph with max/min pooling aggregation, while guaranteeing that all the private data involved in computation still stays at local data holders. To further enhancing the data privacy, a secure pooling aggregation mechanism is proposed. Theoretical and experimental results show that the proposed model achieves the same accuracy as the one learned over the combined data.