Value at Adversarial Risk: A Graph Defense Strategy against Cost-Aware Attacks
DOI:
https://doi.org/10.1609/aaai.v38i12.29282Keywords:
ML: Adversarial Learning & Robustness, ML: Graph-based Machine LearningAbstract
Deep learning methods on graph data have achieved remarkable efficacy across a variety of real-world applications, such as social network analysis and transaction risk detection. Nevertheless, recent studies have illuminated a concerning fact: even the most expressive Graph Neural Networks (GNNs) are vulnerable to graph adversarial attacks. While several methods have been proposed to enhance the robustness of GNN models against adversarial attacks, few have focused on a simple yet realistic approach: valuing the adversarial risks and focused safeguards at the node level. This empowers defenders to allocate heightened security level to vulnerable nodes, while lower to robust nodes. With this new perspective, we propose a novel graph defense strategy RisKeeper, such that the adversarial risk can be directly kept in the input graph. We start at valuing the adversarial risk, by introducing a cost-aware projected gradient descent attack that takes into account both cost avoidance and compliance with costs budgets. Subsequently, we present a learnable approach to ascertain the ideal security level for each individual node by solving a bi-level optimization problem. Through extensive experiments on four real-world datasets, we demonstrate that our method achieves superior performance surpassing state-of-the-art methods. Our in-depth case studies provide further insights into vulnerable and robust structural patterns, serving as inspiration for practitioners to exercise heightened vigilance.Downloads
Published
2024-03-24
How to Cite
Liao, J., Fu, W., Wang, C., Wei, Z., & Xu, J. (2024). Value at Adversarial Risk: A Graph Defense Strategy against Cost-Aware Attacks. Proceedings of the AAAI Conference on Artificial Intelligence, 38(12), 13763-13771. https://doi.org/10.1609/aaai.v38i12.29282
Issue
Section
AAAI Technical Track on Machine Learning III