arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2403.07943cs.LGcs.CR
收藏

重新审视图神经网络中的边扰动:图数据增强与攻击

Revisiting Edge Perturbation for Graph Neural Network in Graph Data Augmentation and Attack

Institute of Computing Technology, Chinese Academy of Sciences(中国科学院计算技术研究所);University of Chinese Academy of Sciences(中国科学院大学);Renmin University of China(中国人民大学);Zhongguancun Laboratory(中关村实验室);Griffith University(格里菲斯大学)

Xin Liu, Yuxiang Zhang, Meng Wu, Mingyu Yan, Kun He, Wei Yan, Shirui Pan, Xiaochun Ye, Dongrui Fan

AI总结:

本文提出统一公式并设计边优先级检测器(EPD),明确图数据增强与攻击中边扰动的界限,实现灵活且高效的图神经网络性能优化。

摘要:

边扰动是修改图结构的基本方法。根据其对图神经网络(GNNs)性能的影响,可将其分为两类:图数据增强和图攻击。令人惊讶的是,这两类边扰动方法采用相同的操作,却对GNNs的准确性产生相反的效果。这些方法在使用边扰动时的明确界限从未被清晰定义。因此,不恰当的扰动可能导致不良结果,需要精确调整以达到预期效果。因此,“为什么边扰动具有双面效应?”以及“是什么使边扰动灵活且有效?”这些问题仍然悬而未决。在本文中,我们将通过提出统一公式并建立两类边扰动方法之间的明确界限来回答这些问题。具体而言,我们进行实验以阐明这些方法之间的差异和相似性,并通过将其工作流程转化为一个优化问题在理论上统一这些方法。然后,我们设计了边优先级检测器(EPD)来生成一种新颖的优先级度量,在流程中桥接这些方法。实验表明,EPD可以灵活地进行增强或攻击,并以更少的时间开销达到与其他方法相当或更优的性能。

英文摘要:

Edge perturbation is a basic method to modify graph structures. It can be categorized into two veins based on their effects on the performance of graph neural networks (GNNs), i.e., graph data augmentation and attack. Surprisingly, both veins of edge perturbation methods employ the same operations, yet yield opposite effects on GNNs' accuracy. A distinct boundary between these methods in using edge perturbation has never been clearly defined. Consequently, inappropriate perturbations may lead to undesirable outcomes, necessitating precise adjustments to achieve desired effects. Therefore, questions of ``why edge perturbation has a two-faced effect?'' and ``what makes edge perturbation flexible and effective?'' still remain unanswered. In this paper, we will answer these questions by proposing a unified formulation and establishing a clear boundary between two categories of edge perturbation methods. Specifically, we conduct experiments to elucidate the differences and similarities between these methods and theoretically unify the workflow of these methods by casting it to one optimization problem. Then, we devise Edge Priority Detector (EPD) to generate a novel priority metric, bridging these methods up in the workflow. Experiments show that EPD can make augmentation or attack flexibly and achieve comparable or superior performance to other counterparts with less time overhead.

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。

↑