Adversarial Inter-Group Link Injection Degrades the Fairness of Graph Neural Networks
arXiv:2209.05957 · doi:10.1109/ICDM54844.2022.00117
Abstract
We present evidence for the existence and effectiveness of adversarial attacks on graph neural networks (GNNs) that aim to degrade fairness. These attacks can disadvantage a particular subgroup of nodes in GNN-based node classification, where nodes of the underlying network have sensitive attributes, such as race or gender. We conduct qualitative and experimental analyses explaining how adversarial link injection impairs the fairness of GNN predictions. For example, an attacker can compromise the fairness of GNN-based node classification by injecting adversarial links between nodes belonging to opposite subgroups and opposite class labels. Our experiments on empirical datasets demonstrate that adversarial fairness attacks can significantly degrade the fairness of GNN predictions (attacks are effective) with a low perturbation rate (attacks are efficient) and without a significant drop in accuracy (attacks are deceptive). This work demonstrates the vulnerability of GNN models to adversarial fairness attacks. We hope our findings raise awareness about this issue in our community and lay a foundation for the future development of GNN models that are more robust to such attacks.
arXiv commentA shorter version of this work has been accepted by IEEE ICDM 2022
References in corpus (8)
- A Comprehensive Survey on Graph Neural Networks
- Equality of Opportunity in Supervised Learning
- Adversarial Attacks on Neural Networks for Graph Data
- Adversarial Attacks on Graph Neural Networks via Meta Learning
- GNNGuard: Defending Graph Neural Networks against Adversarial Attacks
- FairDrop: Biased Edge Dropout for Enhancing Fairness in Graph Representation Learning
- Attacking Graph Convolutional Networks via Rewiring
- Adversarial Attacks and Defenses on Graphs: A Review, A Tool and Empirical Studies