arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2409.19601cs.CR
收藏

黑暗中的内斗:联邦学习中的多标签后门攻击

Infighting in the Dark: Multi-Label Backdoor Attack in Federated Learning

Ye Li, Yanchao Zhao, Chengcheng Zhu, Jiale Zhang

AI总结:

针对联邦学习中非合作多目标的多标签后门攻击场景下现有方法失效的问题,提出Mirage策略,通过分布内后门映射与约束优化实现高效持久攻击,性能优于现有方法且可绕过主流防御。

摘要:

联邦学习(Federated Learning, FL)是一种保护隐私的去中心化机器学习框架,已有研究表明其易受后门攻击。当前研究主要聚焦于单标签后门攻击(Single-Label Backdoor Attack, SBA),这类攻击中攻击者共享一致的目标。然而一个关键事实被忽略:攻击者可能是非合作的,拥有不同目标且独立行动,这构成了一种更具现实性的场景,即多标签后门攻击(Multi-Label Backdoor Attack, MBA)。遗憾的是,现有方法在MBA场景下效果不佳,因为非合作攻击者会相互排斥。在本研究中,我们深入探究并揭示了这种排斥的内在约束:不同目标会构建出相似的后门映射,导致后门功能之间产生冲突。为解决这一局限,我们提出了Mirage——联邦学习中首个非合作MBA策略,它通过构建分布内(in-distribution, ID)后门映射,让攻击者无需串通就能向全局模型注入有效且持久的后门。具体而言,我们提出一种对抗性适配方法,以分布内的方式桥接后门特征与目标分布。此外,我们进一步利用约束优化方法,确保分布内映射能在全局训练动态中存续。大量评估表明,Mirage性能优于多种现有最先进的攻击方法,且能绕过现有防御机制,平均攻击成功率(ASR)超过97%,在900轮训练后仍保持在90%以上。本研究旨在提醒研究人员关注这一潜在威胁,并推动有效防御机制的设计。相关代码已开源。

英文摘要:

Federated Learning (FL), a privacy-preserving decentralized machine learning framework, has been shown to be vulnerable to backdoor attacks. Current research primarily focuses on the Single-Label Backdoor Attack (SBA), wherein adversaries share a consistent target. However, a critical fact is overlooked: adversaries may be non-cooperative, have distinct targets, and operate independently, which exhibits a more practical scenario called Multi-Label Backdoor Attack (MBA). Unfortunately, prior works are ineffective in the MBA scenario since non-cooperative attackers exclude each other. In this work, we conduct an in-depth investigation to uncover the inherent constraints of the exclusion: similar backdoor mappings are constructed for different targets, resulting in conflicts among backdoor functions. To address this limitation, we propose Mirage, the first non-cooperative MBA strategy in FL that allows attackers to inject effective and persistent backdoors into the global model without collusion by constructing in-distribution (ID) backdoor mapping. Specifically, we introduce an adversarial adaptation method to bridge the backdoor features and the target distribution in an ID manner. Additionally, we further leverage a constrained optimization method to ensure the ID mapping survives in the global training dynamics. Extensive evaluations demonstrate that Mirage outperforms various state-of-the-art attacks and bypasses existing defenses, achieving an average ASR greater than 97\% and maintaining over 90\% after 900 rounds. This work aims to alert researchers to this potential threat and inspire the design of effective defense mechanisms. Code has been made open-source.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑