Let Graph be the Go Board: Gradient-free Node Injection Attack for Graph Neural Networks via Reinforcement Learning

TL;DR

This paper introduces G2A2C, a reinforcement learning-based method for black-box node injection attacks on graph neural networks, which does not require surrogate models and effectively exploits vulnerabilities.

Contribution

Proposes G2A2C, a gradient-free, reinforcement learning framework for black-box node injection attacks that bypasses the need for surrogate models and maintains high attack efficacy.

Findings

G2A2C outperforms existing attack methods on multiple benchmarks.

The method requires limited query budgets for successful attacks.

G2A2C maintains similar node feature distributions to original graphs.

Abstract

Graph Neural Networks (GNNs) have drawn significant attentions over the years and been broadly applied to essential applications requiring solid robustness or vigorous security standards, such as product recommendation and user behavior modeling. Under these scenarios, exploiting GNN's vulnerabilities and further downgrading its performance become extremely incentive for adversaries. Previous attackers mainly focus on structural perturbations or node injections to the existing graphs, guided by gradients from the surrogate models. Although they deliver promising results, several limitations still exist. For the structural perturbation attack, to launch a proposed attack, adversaries need to manipulate the existing graph topology, which is impractical in most circumstances. Whereas for the node injection attack, though being more practical, current approaches require training surrogate models to simulate a white-box setting, which results in significant performance downgrade when the surrogate architecture diverges from the actual victim model. To bridge these gaps, in this paper, we study the problem of black-box node injection attack, without training a potentially misleading surrogate model. Specifically, we model the node injection attack as a Markov decision process and propose Gradient-free Graph Advantage Actor Critic, namely G2A2C, a reinforcement learning framework in the fashion of advantage actor critic. By directly querying the victim model, G2A2C learns to inject highly malicious nodes with extremely limited attacking budgets, while maintaining a similar node feature distribution. Through our comprehensive experiments over eight acknowledged benchmark datasets with different characteristics, we demonstrate the superior performance of our proposed G2A2C over the existing state-of-the-art attackers. Source code is publicly available at: https://github.com/jumxglhf/G2A2C}.