Black-box Detection of Backdoor Attacks with Limited Information and Data-Reference-Cited by-同舟云学术

Black-box Detection of Backdoor Attacks with Limited Information and Data

Published:2021-10 Issue: Volume: Page:
ISSN:
Container-title:2021 IEEE/CVF International Conference on Computer Vision (ICCV)
language:
Short-container-title:

Author:

Dong Yinpeng¹,Yang Xiao¹,Deng Zhijie¹,Pang Tianyu¹,Xiao Zihao²,Su Hang¹,Zhu Jun¹

Affiliation:

1. Tsinghua-Bosch Joint ML Center,Institute for AI,Dept. of Comp. Sci. and Tech.

2. Pazhou Lab,Guangzhou,China,510330

Funder

National Key Research and Development Program of China

Publisher

IEEE

Link

Reference50 articles.

4. Fine-pruning: Defending against backdooring attacks on deep neural networks;liu;International Symposium on Research in Attacks Intrusions and Defenses,2018

5. Backdoor learning: A survey;li,2020

Cited by 40 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

2. OCGEC: One-class Graph Embedding Classification for DNN Backdoor Detection;2024 International Joint Conference on Neural Networks (IJCNN);2024-06-30

3. Universal Soldier: Using Universal Adversarial Perturbations for Detecting Backdoor Attacks;2024 54th Annual IEEE/IFIP International Conference on Dependable Systems and Networks Workshops (DSN-W);2024-06-24

4. Backdoor Breakthrough;Advances in Medical Technologies and Clinical Practice;2024-04-12