Near-Black-Box Adversarial Attacks on Graph Neural Networks as An Influence Maximization ProblemDownload PDF

Sep 28, 2020 (edited Mar 05, 2021)ICLR 2021 Conference Blind SubmissionReaders: Everyone
  • Reviewed Version (pdf): https://openreview.net/references/pdf?id=qG9qPyBvDp
  • Keywords: Graph Neural Network, Adversarial Attack, Influence Maximization
  • Abstract: Graph neural networks (GNNs) have attracted increasing interests. With broad deployments of GNNs in real-world applications, there is an urgent need for understanding the robustness of GNNs under adversarial attacks, especially in realistic setups. In this work, we study the problem of attacking GNNs in a restricted near-black-box setup, by perturbing the features of a small set of nodes, with no access to model parameters and model predictions. Our formal analysis draws a connection between this type of attacks and an influence maximization problem on the graph. This connection not only enhances our understanding on the problem of adversarial attack on GNNs, but also allows us to propose a group of effective near-black-box attack strategies. Our experiments verify that the proposed strategies significantly degrade the performance of three popular GNN models and outperform baseline adversarial attack strategies.
  • One-sentence Summary: We establish a connection between adversarial attack on graph neural networks and the influence maximization problem and propose a group of effective black-box attack strategies based on this connection.
  • Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics
21 Replies

Loading