AmI（攻击遇见可解释性）对对抗样本是否鲁棒？
机器学习
2019-02-07 v1 人工智能
密码学与安全
机器学习
摘要
否。
引用
@article{arxiv.1902.02322,
title = {Is AmI (Attacks Meet Interpretability) Robust to Adversarial Examples?},
author = {Nicholas Carlini},
journal= {arXiv preprint arXiv:1902.02322},
year = {2019}
}