Certifiable Robustness to Graph Perturbations

Abstract

Despite the exploding interest in graph neural networks there has been little effort to verify and improve their robustness. This is even more alarming given recent findings showing that they are extremely vulnerable to adversarial attacks on both the graph structure and the node attributes. We propose the first method for verifying certifiable (non-)robustness to graph perturbations for a general class of models that includes graph neural networks and label/feature propagation. By exploiting connections to PageRank and Markov decision processes our certificates can be efficiently (and under many threat models exactly) computed. Furthermore, we investigate robust training procedures that increase the number of certifiably robust nodes while maintaining or improving the clean predictive accuracy.

33rd Conference on Neural Information Processing Systems (NeurIPS 2019), Vancouver, Canada

Cited by in corpus (25)

• Adversarial Attack and Defense on Graph Data: A Survey
• Stability Properties of Graph Neural Networks
• Graph Neural Networks: Taxonomy, Advances and Trends
• GNNGuard: Defending Graph Neural Networks against Adversarial Attacks
• A Comprehensive Survey on Trustworthy Graph Neural Networks: Privacy, Robustness, Fairness, and Explainability
• Single Node Injection Attack against Graph Neural Networks
• Adversarial Attacks and Defenses on Graphs: A Review, A Tool and Empirical Studies
• Stealing Links from Graph Neural Networks
• Robust Counterfactual Explanations on Graph Neural Networks
• How does Heterophily Impact the Robustness of Graph Neural Networks? Theoretical Connections and Practical Implications
• Dissecting the Diffusion Process in Linear Graph Convolutional Networks
• Efficient Robustness Certificates for Discrete Data: Sparsity-Aware Randomized Smoothing for Graphs, Images and More
• Decentralized Inference with Graph Neural Networks in Wireless Communication Systems
• Certified Robustness of Community Detection against Adversarial Structural Perturbation via Randomized Smoothing
• GraphAttacker: A General Multi-Task GraphAttack Framework
• Evaluating Robustness of Predictive Uncertainty Estimation: Are Dirichlet-based Models Reliable?
• Certified Robustness of Graph Neural Networks against Adversarial Structural Perturbation
• Interpretable Stability Bounds for Spectral Graph Filters
• Understanding Structural Vulnerability in Graph Convolutional Networks
• On the Stability of Graph Convolutional Neural Networks under Edge Rewiring
• Topological Effects on Attacks Against Vertex Classification
• AN-GCN: An Anonymous Graph Convolutional Network Defense Against Edge-Perturbing Attack
• Graphs for deep learning representations
• Certified Robustness of Graph Classification against Topology Attack with Randomized Smoothing
• Spatio-Temporal Sparsification for General Robust Graph Convolution Networks