Sajadmanesh, SinaGatica-Perez, Daniel2022-05-232022-05-232022-05-232021-01-0110.1145/3460120.3484565https://infoscience.epfl.ch/handle/20.500.14299/187952WOS:000768478302010Graph Neural Networks (GNNs) have demonstrated superior performance in learning node representations for various graph inference tasks. However, learning over graph data can raise privacy concerns when nodes represent people or human-related variables that involve sensitive or personal information. In this paper, we study the problem of node data privacy, where graph nodes (e.g., social network users) have potentially sensitive data that is kept private, but they could be beneficial for a central server for training a GNN over the graph. To address this problem, we propose a privacy-preserving, architecture-agnostic GNN learning framework with formal privacy guarantees based on Local Differential Privacy (LDP). Specifically, we develop a locally private mechanism to perturb and compress node features, which the server can efficiently collect to approximate the GNN's neighborhood aggregation step. Furthermore, to improve the accuracy of the estimation, we prepend to the GNN a denoising layer, called KProp, which is based on the multi-hop aggregation of node features. Finally, we propose a robust algorithm for learning with privatized noisy labels, where we again benefit from KProp's denoising capability to increase the accuracy of label inference for node classification. Extensive experiments conducted over real-world datasets demonstrate that our method can maintain a satisfying level of accuracy with low privacy loss.Computer Science, Information SystemsComputer Science, Theory & MethodsTelecommunicationsComputer Sciencedifferential privacyprivate learninggraph neural networksnode classificationtext::conference output::conference proceedings::conference paper