GraphAttacker: A General Multi-Task GraphAttack Framework

TL;DR

GraphAttacker is a versatile GAN-based framework that generates diverse, effective adversarial attacks on various graph neural network tasks, enhancing robustness testing and understanding of GNN vulnerabilities.

Contribution

It introduces a flexible, multi-strategy attack framework adaptable to different graph analysis tasks, with a novel similarity modification rate for stealthier attacks.

Findings

Achieves state-of-the-art attack performance across multiple tasks.

Effective against models with and without adversarial training.

Analyzes task-specific attack characteristics.

Abstract

Graph neural networks (GNNs) have been successfully exploited in graph analysis tasks in many real-world applications. The competition between attack and defense methods also enhances the robustness of GNNs. In this competition, the development of adversarial training methods put forward higher requirement for the diversity of attack examples. By contrast, most attack methods with specific attack strategies are difficult to satisfy such a requirement. To address this problem, we propose GraphAttacker, a novel generic graph attack framework that can flexibly adjust the structures and the attack strategies according to the graph analysis tasks. GraphAttacker generates adversarial examples through alternate training on three key components: the multi-strategy attack generator (MAG), the similarity discriminator (SD), and the attack discriminator (AD), based on the generative adversarial network (GAN). Furthermore, we introduce a novel similarity modification rate SMR to conduct a stealthier attack considering the change of node similarity distribution. Experiments on various benchmark datasets demonstrate that GraphAttacker can achieve state-of-the-art attack performance on graph analysis tasks of node classification, graph classification, and link prediction, no matter the adversarial training is conducted or not. Moreover, we also analyze the unique characteristics of each task and their specific response in the unified attack framework. The project code is available at https://github.com/honoluluuuu/GraphAttacker.