Search Results for author:
Found 2 papers, 0 papers with code
Defense-guided Transferable Adversarial Attacks
Experimentally, we show that our ASR of adversarial attack reaches to 58. 38% on average, which outperforms the state-of-the-art method by 12. 1% on the normally trained models and by 11. 13% on the adversarially trained models.
AdvJND: Generating Adversarial Examples with Just Noticeable Difference
To alleviate the tradeoff between the attack success rate and image fidelity, we propose a method named AdvJND, adding visual model coefficients, just noticeable difference coefficients, in the constraint of a distortion function when generating adversarial examples.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
