Information Obfuscation of Graph Neural Networks

Abstract

While the advent of Graph Neural Networks (GNNs) has greatly improved node and graph representation learning in many applications, the neighborhood aggregation scheme exposes additional vulnerabilities to adversaries seeking to extract node-level information about sensitive attributes. In this paper, we study the problem of protecting sensitive attributes by information obfuscation when learning with graph structured data. We propose a framework to locally filter out pre-determined sensitive attributes via adversarial training with the total variation and the Wasserstein distance. Our method creates a strong defense against inference attacks, while only suffering small loss in task performance. Theoretically, we analyze the effectiveness of our framework against a worst-case adversary, and characterize an inherent trade-off between maximizing predictive accuracy and minimizing information leakage. Experiments across multiple datasets from recommender systems, knowledge graphs and quantum chemistry demonstrate that the proposed approach provides a robust defense across various graph structures and tasks, while producing competitive GNN encoders for downstream tasks.

arXiv commentICML 2021; Code is available at https://github.com/liaopeiyuan/GAL

References in corpus (19)

• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• The information bottleneck method
• Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering
• Fast Graph Representation Learning with PyTorch Geometric
• Graph Convolutional Matrix Completion
• Representation Learning on Graphs with Jumping Knowledge Networks
• Convolutional Networks on Graphs for Learning Molecular Fingerprints
• Deep Variational Information Bottleneck
• Adversarial Attacks on Node Embeddings via Graph Poisoning
• Adversarial Attack on Graph Structured Data
• Solving a Class of Non-Convex Min-Max Games Using Iterative First Order Methods
• Graph Neural Tangent Kernel: Fusing Graph Neural Networks with Graph Kernels
• What Can Neural Networks Reason About?
• Compositional Fairness Constraints for Graph Embeddings
• GraphNorm: A Principled Approach to Accelerating Graph Neural Network Training
• Attacking Graph Convolutional Networks via Rewiring
• Just Jump: Dynamic Neighborhood Aggregation in Graph Neural Networks

Cited by in corpus (2)

• Adversarial Attack and Defense on Graph Data: A Survey
• Single-Node Attacks for Fooling Graph Neural Networks