arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2006.05057cs.LGstat.ML
收藏

面向更实用的图神经网络对抗攻击

Towards More Practical Adversarial Attacks on Graph Neural Networks

University of Michigan(密歇根大学)

Jiaqi Ma, Shuangrui Ding, Qiaozhu Mei

摘要:

我们研究了在一个新颖而现实的约束下对图神经网络(GNNs)的黑盒攻击:攻击者只能访问网络中的一个节点子集,并且他们只能攻击其中的一小部分。在此设置下,节点选择步骤是必不可少的。我们证明了GNN模型的结构归纳偏差可以成为这种攻击的有效来源。具体来说,通过利用GNN的反向传播和随机游走之间的联系,我们证明了常见的基于梯度的白盒攻击可以通过梯度和类似于PageRank的重要性得分之间的联系推广到黑盒设置。在实践中,我们发现基于此重要性分数的攻击确实会大幅增加分类损失,但它们未能显著增加错误分类率。我们的理论和实证分析表明,有一个损失和错误分类率之间的差异,因为后者提出了一个递减返回模式时,攻击节点的数量增加。因此,我们提出了一个贪婪的过程来纠正的重要性分数,考虑到递减返回模式。实验结果表明,该方法可以显着提高常见GNN对真实世界数据的误分类率,而无需访问模型参数或预测。

英文摘要:

We study the black-box attacks on graph neural networks (GNNs) under a novel and realistic constraint: attackers have access to only a subset of nodes in the network, and they can only attack a small number of them. A node selection step is essential under this setup. We demonstrate that the structural inductive biases of GNN models can be an effective source for this type of attacks. Specifically, by exploiting the connection between the backward propagation of GNNs and random walks, we show that the common gradient-based white-box attacks can be generalized to the black-box setting via the connection between the gradient and an importance score similar to PageRank. In practice, we find attacks based on this importance score indeed increase the classification loss by a large margin, but they fail to significantly increase the mis-classification rate. Our theoretical and empirical analyses suggest that there is a discrepancy between the loss and mis-classification rate, as the latter presents a diminishing-return pattern when the number of attacked nodes increases. Therefore, we propose a greedy procedure to correct the importance score that takes into account of the diminishing-return pattern. Experimental results show that the proposed procedure can significantly increase the mis-classification rate of common GNNs on real-world data without access to model parameters nor predictions.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑