Data for deep learning should be protected for privacy preserving. Researchers have come up with the notion of learnable image encryption to satisfy the requirement. However, existing privacy preserving approaches have never considered the threat of adversarial attacks. In this paper, we ran an adversarial test on learnable image encryption in five different scenarios. The results show different behaviors of the network in the variable key scenarios and suggest learnable image encryption provides certain level of adversarial robustness.
@article{arxiv.1907.13342,
title = {Adversarial Test on Learnable Image Encryption},
author = {MaungMaung AprilPyone and Warit Sirichotedumrong and Hitoshi Kiya},
journal= {arXiv preprint arXiv:1907.13342},
year = {2019}
}
Comments
To be appeared in 2019 IEEE 8th Global Conference on Consumer Electronics (GCCE 2019)