基于子图采样的大规模图对抗性攻击方法
Subgraph sampling-based adversarial attack method for large-scale graphs
高昕 1安冬冬 1章晓峰2
作者信息
- 1. 上海师范大学 信息与机电工程学院,上海 201418
- 2. 上海新致软件股份有限公司,上海 200120
- 折叠
摘要
为提高对抗性攻击在大规模图上的攻击效率,提出了基于子图采样的对抗样本生成方法.该方法通过引入PageRank、余弦相似度及K跳子图等技术,提取与目标节点高度相关的子图,在大规模图上缓解了计算梯度效率较低的问题,在降低被攻击模型准确性的同时提升了攻击的隐蔽性.实验结果表明:所提出的对抗性攻击方法与基于梯度攻击的GradArgmax算法相比,在Cora数据集上提升了 30.7%的攻击性能,且在Reddit大规模数据上能够计算GradArgmax算法无法计算的攻击扰动.
Abstract
A subgraph sampling-based adversarial example generation method was proposed to enhance the efficiency of adversarial attacks on large-scale graphs.PageRank,cosine similarity,and K-hop subgraphs were employed to extract subgraphs highly relevant to the target node in this method,alleviating the issue of low gradient computation efficiency in large-scale graphs.The stealthiness of the attack was also increased while reducing the accuracy of the attacked model.Experimental results showed that attack performance was improved by 30.7%on the Cora dataset by this adversarial attack method compared to the GradArgmax algorithm,and attack perturbations on large-scale like Reddit dataset could be computed which the GradArgmax algorithm could not achieve.
关键词
图神经网络/对抗性攻击/子图提取算法Key words
graph neural network/adversarial attack/subgraph extraction algorithm引用本文复制引用
基金项目
国家自然科学基金青年基金(62302308)
上海市青年科技英才扬帆计划(21YF1432900)
出版年
2024
NETL
NSTL
万方数据