Robustness of Graph Neural Networks at Scale

Simon Geisler, Tobias Schmidt, Hakan Şirin, Daniel Zügner, Aleksandar Bojchevski, Stephan Günnemann

Keywords: Adversarial Robustness, Graph Neural Networks, Scalability, Adversarial Defenses, Adversarial Attacks

TL;DR: We evaluate adversarial (non-) robustness of Graph Neural Networks on graphs up to three orders of magnitude larger than before.

Abstract: Graph Neural Networks (GNNs) are increasingly important given their popularity and the diversity of applications. Yet, existing studies of their vulnerability to adversarial attacks rely on relatively small graphs. We address this gap and study how to attack and defend GNNs at scale. We propose two sparsity-aware first-order optimization attacks that maintain an efficient representation despite optimizing over a number of parameters which is quadratic in the number of nodes. We show that common surrogate losses are not well-suited for global attacks on GNNs. Our alternatives can double the attack strength. Moreover, to improve GNNs' reliability we design a robust aggregation function, Soft Median, resulting in an effective defense at all scales. We evaluate our attacks and defense with standard GNNs on graphs more than 100 times larger compared to previous work. We even scale one order of magnitude further by extending our techniques to a scalable GNN.

Code Of Conduct: I certify that all co-authors of this work have read and commit to adhering to the NeurIPS Statement on Ethics, Fairness, Inclusivity, and Code of Conduct.

Code: https://www.in.tum.de/daml/robustness-of-gnns-at-scale/

Community Implementations: [ 1 code implementation](https://www.catalyzex.com/paper/arxiv:2110.14038/code)