Power up! Robust Graph Convolutional Network based on Graph Powering

Ming Jin, Heng Chang, Wenwu Zhu, Somayeh Sojoudi

Keywords: graph mining, graph neural network, adversarial robustness

TL;DR: We propose a framework for robust graph neural networks based on graph powering

Abstract: Graph convolutional networks (GCNs) are powerful tools for graph-structured data. However, they have been recently shown to be vulnerable to topological attacks. To enhance adversarial robustness, we go beyond spectral graph theory to robust graph theory. By challenging the classical graph Laplacian, we propose a new convolution operator that is provably robust in the spectral domain and is incorporated in the GCN architecture to improve expressivity and interpretability. By extending the original graph to a sequence of graphs, we also propose a robust training paradigm that encourages transferability across graphs that span a range of spatial and spectral characteristics. The proposed approaches are demonstrated in extensive experiments to {simultaneously} improve performance in both benign and adversarial situations.

Code: https://www.dropbox.com/sh/p36pzx1ock2iamo/AABEr7FtM5nqwC4i9nICLIsta?dl=0