Node Feature Kernels Increase Graph Convolutional Network Robustness

Abstract

The robustness of the much-used Graph Convolutional Networks (GCNs) to perturbations of their input is becoming a topic of increasing importance. In this paper, the random GCN is introduced for which a random matrix theory analysis is possible. This analysis suggests that if the graph is sufficiently perturbed, or in the extreme case random, then the GCN fails to benefit from the node features. It is furthermore observed that enhancing the message passing step in GCNs by adding the node feature kernel to the adjacency matrix of the graph structure solves this problem. An empirical study of a GCN utilised for node classification on six real datasets further confirms the theoretical findings and demonstrates that perturbations of the graph structure can result in GCNs performing significantly worse than Multi-Layer Perceptrons run on the node features alone. In practice, adding a node feature kernel to the message passing of perturbed graphs results in a significant improvement of the GCN's performance, thereby rendering it more robust to graph perturbations. Our code is publicly available at:https://github.com/ChangminWu/RobustGCN.

arXiv comment16 pages, 5 figures

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Stochastic blockmodels and community structure in networks
• Fast Graph Representation Learning with PyTorch Geometric
• Pitfalls of Graph Neural Network Evaluation
• Adversarial Attacks on Graph Neural Networks via Meta Learning
• Adversarial Attack and Defense on Graph Data: A Survey
• Graph Neural Networks: Taxonomy, Advances and Trends
• Concentration of Measure and Large Random Matrices with an application to Sample Covariance Matrices
• Provably Powerful Graph Networks
• Power up! Robust Graph Convolutional Network via Graph Powering