Adversarial Attack on Hierarchical Graph Pooling Neural Networks

TL;DR

This paper investigates the vulnerability of hierarchical graph pooling neural networks to adversarial attacks, proposing a new attack framework and demonstrating its transferability and the effectiveness of robust training for defense.

Contribution

It introduces the first adversarial attack framework targeting hierarchical GNNs for graph classification, highlighting their robustness issues and proposing a defense strategy.

Findings

Adversarial samples can successfully fool hierarchical GNNs.

Transferability of adversarial samples across models is high.

Robust training improves model resistance to attacks.

Abstract

Recent years have witnessed the emergence and development of graph neural networks (GNNs), which have been shown as a powerful approach for graph representation learning in many tasks, such as node classification and graph classification. The research on the robustness of these models has also started to attract attentions in the machine learning field. However, most of the existing work in this area focus on the GNNs for node-level tasks, while little work has been done to study the robustness of the GNNs for the graph classification task. In this paper, we aim to explore the vulnerability of the Hierarchical Graph Pooling (HGP) Neural Networks, which are advanced GNNs that perform very well in the graph classification in terms of prediction accuracy. We propose an adversarial attack framework for this task. Specifically, we design a surrogate model that consists of convolutional and pooling operators to generate adversarial samples to fool the hierarchical GNN-based graph classification models. We set the preserved nodes by the pooling operator as our attack targets, and then we perturb the attack targets slightly to fool the pooling operator in hierarchical GNNs so that they will select the wrong nodes to preserve. We show the adversarial samples generated from multiple datasets by our surrogate model have enough transferability to attack current state-of-art graph classification models. Furthermore, we conduct the robust train on the target models and demonstrate that the retrained graph classification models are able to better defend against the attack from the adversarial samples. To the best of our knowledge, this is the first work on the adversarial attack against hierarchical GNN-based graph classification models.