Blink: Link Local Differential Privacy in Graph Neural Networks via Bayesian Estimation

TL;DR

This paper introduces Blink, a method for training graph neural networks with link local differential privacy using Bayesian estimation, improving accuracy while preserving privacy in decentralized settings.

Contribution

We propose a novel link local differential privacy mechanism for GNNs that uses Bayesian estimation to better denoise graph topology, with multiple variants and a hybrid approach for different privacy levels.

Findings

Outperforms existing methods in accuracy under various privacy budgets

Effectively bounds mean absolute error of inferred link probabilities

Demonstrates robustness of the hybrid mechanism across privacy settings

Abstract

Graph neural networks (GNNs) have gained an increasing amount of popularity due to their superior capability in learning node embeddings for various graph inference tasks, but training them can raise privacy concerns. To address this, we propose using link local differential privacy over decentralized nodes, enabling collaboration with an untrusted server to train GNNs without revealing the existence of any link. Our approach spends the privacy budget separately on links and degrees of the graph for the server to better denoise the graph topology using Bayesian estimation, alleviating the negative impact of LDP on the accuracy of the trained GNNs. We bound the mean absolute error of the inferred link probabilities against the ground truth graph topology. We then propose two variants of our LDP mechanism complementing each other in different privacy settings, one of which estimates fewer links under lower privacy budgets to avoid false positive link estimates when the uncertainty is high, while the other utilizes more information and performs better given relatively higher privacy budgets. Furthermore, we propose a hybrid variant that combines both strategies and is able to perform better across different privacy budgets. Extensive experiments show that our approach outperforms existing methods in terms of accuracy under varying privacy budgets.