Explainability-Based Adversarial Attack on Graphs Through Edge Perturbation

2023/12/28 by Dibaloke Chanda, Chanda, Dibaloke, Saba Heidari Gheshlaghi +3
Computer Science · #Advanced Graph Neural Networks #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #Explainable Artificial Intelligence (XAI) #FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · pdf · doi:10.48550/arxiv.2312.17301

openalex publication_date 2023/12/28 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Despite the success of graph neural networks (GNNs) in various domains, they exhibit susceptibility to adversarial attacks. Understanding these vulnerabilities is crucial for developing robust and secure applications. In this paper, we investigate the impact of test time adversarial attacks through edge perturbations which involve both edge insertions and deletions. A novel explainability-based method is proposed to identify important nodes in the graph and perform edge perturbation between these nodes. The proposed method is tested for node classification with three different architectures and datasets. The results suggest that introducing edges between nodes of different classes has higher impact as compared to removing edges among nodes within the same class.

Related

• Adversarial Attacks and Defenses on Graphs: A Review, A Tool and Empirical Studies
• EDoG: Adversarial Edge Detection For Graph Neural Networks
• Unifying Adversarial Perturbation for Graph Neural Networks
• Provably Robust Explainable Graph Neural Networks against Graph Perturbation Attacks
• Graph Neural Network Explanations are Fragile