Backdoor Defense via Adaptively Splitting Poisoned Dataset
Author:
Affiliation:
1. Tsinghua University
2. Tencent Security Zhuque Lab
3. University of Oxford
4. Tencent Security Platform Department
Publisher
IEEE
Link
http://xplorestaging.ieee.org/ielx7/10203037/10203050/10204454.pdf?arnumber=10204454
Reference62 articles.
1. Model-agnostic meta-learning for fast adaptation of deep networks;finn;ICML,2017
2. Solving large scale linear prediction problems using stochastic gradient descent algorithms
3. Robust anomaly detection and backdoor attack detection via differential privacy;du;ICLRE,2020
4. Adversarial unlearning of backdoors via implicit hypergradient;zeng;ICLRE,2022
5. STRIP
Cited by 8 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
1. Adaptive Robust Learning Against Backdoor Attacks in Smart Homes;IEEE Internet of Things Journal;2024-07-01
2. Need for Speed: Taming Backdoor Attacks with Speed and Precision;2024 IEEE Symposium on Security and Privacy (SP);2024-05-19
3. Defending against Clean-Image Backdoor Attack in Multi-Label Classification;ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP);2024-04-14
4. Anti-Backdoor Model: A Novel Algorithm to Remove Backdoors in a Non-Invasive Way;IEEE Transactions on Information Forensics and Security;2024
5. VFLIP: A Backdoor Defense for Vertical Federated Learning via Identification and Purification;Lecture Notes in Computer Science;2024
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3