Poisoning × Evasion: Symbiotic Adversarial Robustness for Graph Neural Networks
Not on the map. Why?
Abstract (arXiv)
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.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 14 of its 24 references that are in the map. First posted on arXiv 9 Dec 2023.
Builds on
14 of 24 references.
References · 14
- 2016Semi-Supervised Classification with Graph Convolutional Networkscited by 11,282
- 2017Graph Attention Networkscited by 8,854
- 2018Predict then Propagate: Graph Neural Networks meet Personalized PageRankcited by 987
- 2020Adaptive Universal Generalized PageRank Graph Neural Networkcited by 592
- 2018Adversarial Attacks on Neural Networks for Graph Datacited by 395
- 2018Adversarial Attack on Graph Structured Datacited by 300
- 2019Topology Attack and Defense for Graph Neural Networks: An Optimization Perspectivecited by 196
- 2018Adversarial Attacks on Node Embeddings via Graph Poisoningcited by 109
- 2021Robustness of Graph Neural Networks at Scalecited by 100
- 2023Are Defenses for Graph Neural Networks Robust?cited by 38
- 2019Data Poisoning Attack against Knowledge Graph Embeddingcited by 34
- 2023Adversarial Training for Graph Neural Networks: Pitfalls, Solutions, and New Directionscited by 28
2 more
Built on by
Not cited yet.