REVIEW 2 cited by
Regional Homogeneity: Towards Learning Transferable Universal Adversarial Perturbations Against Defenses
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
This paper focuses on learning transferable adversarial examples specifically against defense models (models to defense adversarial attacks). In particular, we show that a simple universal perturbation can fool a series of state-of-the-art defenses. Adversarial examples generated by existing attacks are generally hard to transfer to defense models. We observe the property of regional homogeneity in adversarial perturbations and suggest that the defenses are less robust to regionally homogeneous perturbations. Therefore, we propose an effective transforming paradigm and a customized gradient transformer module to transform existing perturbations into regionally homogeneous ones. Without explicitly forcing the perturbations to be universal, we observe that a well-trained gradient transformer module tends to output input-independent gradients (hence universal) benefiting from the under-fitting phenomenon. Thorough experiments demonstrate that our work significantly outperforms the prior art attacking algorithms (either image-dependent or universal ones) by an average improvement of 14.0% when attacking 9 defenses in the transfer-based attack setting. In addition to the cross-model transferability, we also verify that regionally homogeneous perturbations can well transfer across different vision tasks (attacking with the semantic segmentation task and testing on the object detection task). The code is available here: https://github.com/LiYingwei/Regional-Homogeneity.
Forward citations
Cited by 2 Pith papers
-
Transferring Robustness for Graph Neural Network Against Poisoning Attacks
PA-GNN meta-learns to penalize adversarial edges on clean graphs and retains that penalization when fine-tuned on a poisoned graph, improving node classification accuracy under poisoning attacks.
-
Deep Neural Network Ensembles against Deception: Ensemble Diversity, Accuracy and Robustness
Selecting DNN ensemble teams by low Kappa disagreement is presented as a defense against adversarial examples, but the evidence is preliminary and incomplete.
Discussion (0). Continue with ORCID to comment.