1 paper
Yechao Zhang, Shengshan Hu, Leo Yu Zhang +5
Adversarial examples (AEs) for DNNs have been shown to be transferable: AEs that successfully fool white-box surrogate models can also deceive other black-box models with different…