Detecting Topology Attacks against Graph Neural Networks

by   Senrong Xu, et al.
Nanjing University
University of Illinois at Urbana-Champaign
The University of Texas at Dallas

Graph neural networks (GNNs) have been widely used in many real applications, and recent studies have revealed their vulnerabilities against topology attacks. To address this issue, existing efforts have mainly been dedicated to improving the robustness of GNNs, while little attention has been paid to the detection of such attacks. In this work, we study the victim node detection problem under topology attacks against GNNs. Our approach is built upon the key observation rooted in the intrinsic message passing nature of GNNs. That is, the neighborhood of a victim node tends to have two competing group forces, pushing the node classification results towards the original label and the targeted label, respectively. Based on this observation, we propose to detect victim nodes by deliberately designing an effective measurement of the neighborhood variance for each node. Extensive experimental results on four real-world datasets and five existing topology attacks show the effectiveness and efficiency of the proposed detection approach.


page 1

page 2

page 3

page 4


Unnoticeable Backdoor Attacks on Graph Neural Networks

Graph Neural Networks (GNNs) have achieved promising results in various ...

Robust Node Classification on Graphs: Jointly from Bayesian Label Transition and Topology-based Label Propagation

Node classification using Graph Neural Networks (GNNs) has been widely a...

The Devil is in the Conflict: Disentangled Information Graph Neural Networks for Fraud Detection

Graph-based fraud detection has heretofore received considerable attenti...

Graph Neural Networks for Hardware Vulnerability Analysis – Can you Trust your GNN?

The participation of third-party entities in the globalized semiconducto...

Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages

Nowadays, Graph Neural Networks (GNNs) following the Message Passing par...

Graph Agent Network: Empowering Nodes with Decentralized Communications Capabilities for Adversarial Resilience

End-to-end training with global optimization have popularized graph neur...

Unifying gradient regularization for Heterogeneous Graph Neural Networks

Heterogeneous Graph Neural Networks (HGNNs) are a class of powerful deep...

Please sign up or login with your details

Forgot password? Click here to reset