Weiwei Zhao

Improved black-box attack based on query and perturbation distribution

2021

Adversarial examples cause the deep neural network prediction error, which is a great threat to the deep neural network. How to generate more natural adversarial examples and improve the robustness of deep neural networks has …