@article{feinman2017detecting,
author = {Feinman, Reuben and Curtin, Ryan R and Shintre, Saurabh and Gardner, Andrew B},
file = {:E$\backslash$:/GoogleDrive/{\#}BAI TAO{\#} Adversarial ML/Mendeley/2017 - Detecting adversarial samples from artifacts.pdf:pdf},
journal = {arXiv preprint arXiv:1703.00410},
title = {{Detecting adversarial samples from artifacts}},
year = {2017}
}
I am not sure I understand this paper. It seems that this work estimated the probability of the data point being an adversarial example. If got time, I will read it for second time.