ethan-yi-zeng/frequency-backdoor

ICCV 2021, We find most existing triggers of backdoor attacks in deep learning contain severe artifacts in the frequency domain. This Repo. explores how we can use these artifacts to develop stronger backdoor defenses and attacks.

语言：Jupyter Notebook · GitHub 星标：48 · Fork：7

话题：backdoor-attacks · deep-learning · security

在 GitHub 上查看原仓库