Structack: Structure-based Adversarial Attacks on Graph Neural Networks

TL;DR

This paper introduces Structack, a novel structure-based adversarial attack on GNNs that exploits structural properties like centrality and similarity, demonstrating effective performance even with limited information and computational resources.

Contribution

The paper presents Structack, a new uninformed attack method on GNNs that leverages structural graph features, approaching the effectiveness of informed attacks with lower computational cost.

Findings

Structack significantly reduces GNN performance using only structural information.

Uninformed attacks can approach the effectiveness of informed attacks.

Structural properties like centrality and similarity are key to attack success.

Abstract

Recent work has shown that graph neural networks (GNNs) are vulnerable to adversarial attacks on graph data. Common attack approaches are typically informed, i.e. they have access to information about node attributes such as labels and feature vectors. In this work, we study adversarial attacks that are uninformed, where an attacker only has access to the graph structure, but no information about node attributes. Here the attacker aims to exploit structural knowledge and assumptions, which GNN models make about graph data. In particular, literature has shown that structural node centrality and similarity have a strong influence on learning with GNNs. Therefore, we study the impact of centrality and similarity on adversarial attacks on GNNs. We demonstrate that attackers can exploit this information to decrease the performance of GNNs by focusing on injecting links between nodes of low similarity and, surprisingly, low centrality. We show that structure-based uninformed attacks can approach the performance of informed attacks, while being computationally more efficient. With our paper, we present a new attack strategy on GNNs that we refer to as Structack. Structack can successfully manipulate the performance of GNNs with very limited information while operating under tight computational constraints. Our work contributes towards building more robust machine learning approaches on graphs.