English

Improving the Transferability of Adversarial Examples with New Iteration Framework and Input Dropout

Machine Learning 2021-06-24 v2 Computer Vision and Pattern Recognition

Abstract

Deep neural networks(DNNs) is vulnerable to be attacked by adversarial examples. Black-box attack is the most threatening attack. At present, black-box attack methods mainly adopt gradient-based iterative attack methods, which usually limit the relationship between the iteration step size, the number of iterations, and the maximum perturbation. In this paper, we propose a new gradient iteration framework, which redefines the relationship between the above three. Under this framework, we easily improve the attack success rate of DI-TI-MIM. In addition, we propose a gradient iterative attack method based on input dropout, which can be well combined with our framework. We further propose a multi dropout rate version of this method. Experimental results show that our best method can achieve attack success rate of 96.2\% for defense model on average, which is higher than the state-of-the-art gradient-based attacks.

Keywords

Cite

@article{arxiv.2106.01617,
  title  = {Improving the Transferability of Adversarial Examples with New Iteration Framework and Input Dropout},
  author = {Pengfei Xie and Linyuan Wang and Ruoxi Qin and Kai Qiao and Shuhao Shi and Guoen Hu and Bin Yan},
  journal= {arXiv preprint arXiv:2106.01617},
  year   = {2021}
}
R2 v1 2026-06-24T02:46:56.063Z