Attacking Graph Convolutional Networks via Rewiring

TL;DR

This paper introduces a novel adversarial attack method on Graph Neural Networks that uses graph rewiring via reinforcement learning, making perturbations less noticeable while effectively degrading model performance.

Contribution

It proposes a new graph rewiring attack strategy learned through reinforcement learning, offering a less detectable alternative to traditional edge modifications.

Findings

Rewiring-based attacks are effective against GNNs.

The method produces less noticeable perturbations.

Experimental results show significant attack success.

Abstract

Graph Neural Networks (GNNs) have boosted the performance of many graph related tasks such as node classification and graph classification. Recent researches show that graph neural networks are vulnerable to adversarial attacks, which deliberately add carefully created unnoticeable perturbation to the graph structure. The perturbation is usually created by adding/deleting a few edges, which might be noticeable even when the number of edges modified is small. In this paper, we propose a graph rewiring operation which affects the graph in a less noticeable way compared to adding/deleting edges. We then use reinforcement learning to learn the attack strategy based on the proposed rewiring operation. Experiments on real world graphs demonstrate the effectiveness of the proposed framework. To understand the proposed framework, we further analyze how its generated perturbation to the graph structure affects the output of the target model.