Identifying Backdoored Graphs in Graph Neural Network Training: An Explanation-Based Approach with Novel Metrics

TL;DR

This paper introduces a novel explanation-based detection method for backdoor attacks in Graph Neural Networks, utilizing seven new metrics and adaptive attack evaluation on benchmark datasets.

Contribution

The paper presents a new detection approach that leverages graph explanations and seven innovative metrics to identify backdoor attacks in GNNs.

Findings

High detection performance on multiple benchmark datasets

Effective against various attack models

Advances security in GNN applications

Abstract

Graph Neural Networks (GNNs) have gained popularity in numerous domains, yet they are vulnerable to backdoor attacks that can compromise their performance and ethical application. The detection of these attacks is crucial for maintaining the reliability and security of GNN classification tasks, but existing methods are often inflexible, relying on single metrics that fail to capture the full range of backdoor behaviors. Recognizing the challenge in detecting such intrusions, we devised a novel detection method that creatively leverages graph-level explanations. By extracting and transforming secondary outputs from GNN explanation mechanisms, we developed seven innovative metrics for effective detection of backdoor attacks on GNNs. Additionally, we develop an adaptive attack to rigorously evaluate our approach. We test our method on multiple benchmark datasets and examine its efficacy against various attack models. Our results show that our method can achieve high detection performance, marking a significant advancement in safeguarding GNNs against backdoor attacks.