Poisoning $\times$ Evasion: Symbiotic Adversarial Robustness for Graph Neural Networks
This addresses adversarial robustness in GNNs, which is an incremental advance in security for graph-based machine learning applications.
The paper tackles the vulnerability of Graph Neural Networks (GNNs) to adversarial attacks by combining poisoning and evasion threat models, showing that this symbiosis substantially improves attack efficacy with a memory-efficient adaptive end-to-end method.
It is well-known that deep learning models are vulnerable to small input perturbations. Such perturbed instances are called adversarial examples. Adversarial examples are commonly crafted to fool a model either at training time (poisoning) or test time (evasion). In this work, we study the symbiosis of poisoning and evasion. We show that combining both threat models can substantially improve the devastating efficacy of adversarial attacks. Specifically, we study the robustness of Graph Neural Networks (GNNs) under structure perturbations and devise a memory-efficient adaptive end-to-end attack for the novel threat model using first-order optimization.