FL Backdoor Defense (Ⅰ) - FLAME: Taming Backdoors in Federated Learning_“flame: taming backdoors in federated learning,”

作者：天景科技苑 | 2024-07-26 06:51:47

踩

“flame: taming backdoors in federated learning,”

论文笔记 - FLAME: Taming Backdoors in Federated Learning

论文标题	FLAME: Taming Backdoors in Federated Learning
论文作者	Thien Duc Nguyen, Phillip Rieger, Huili Chen, Hossein Yalame, Helen Möllering, Hossein Fereidooni, Samuel Marchal, Markus Miettinen, Azalia Mirhoseini, Shaza Zeitouni, Farinaz Koushanfar, Ahmad-Reza Sadeghi, Thomas Schneider
科研机构	Technical University of Darmstadt, University of California San Diego, Aalto University, Google
录用会议	USENIX Security 2022
摘要概述	现有的联邦学习后门攻击防御算法主要围绕着检测并过滤或者差分隐私两种方法，前者仅考虑了非常具体和有限的恶意客户端。后者破坏了全局模型的性能。对于这两种缺陷作者提出了 FLAME 算法，通过近似要注入噪声的量来实现全局模型后门的消除。具体而言，结合模型异常检测和权重裁剪的方法实现注入噪声的量最小化<

声明：本文内容由网友自发贡献，不代表【wpsshop博客】立场，版权归原作者所有，本站不承担相应法律责任。如您发现有侵权的内容，请联系我们。转载请注明出处：https://www.wpsshop.cn/w/天景科技苑/article/detail/884043