通信学报2024，Vol.45Issue(12)：16-27,12.DOI:10.11959/j.issn.1000-436x.2024264
无人系统中离线强化学习的隐蔽数据投毒攻击方法
Stealthy data poisoning attack method on offline reinforcement learning in unmanned systems
摘要
Abstract
Aiming at the limitations in effectiveness and stealth of existing offline reinforcement learning(RL)data poi-soning attacks,a critical time-step dynamic poisoning attack was proposed,perturbing important samples to achieve effi-cient and covert attacks.Temporal difference errors,identified through theoretical analysis as crucial for model learning,were used to guide poisoning target selection.A bi-objective optimization approach was introduced to minimize perturba-tion magnitude while maximizing the negative impact on performance.Experimental results show that with only a 1%poisoning rate,the method reduces agent performance by 84%,revealing the sensitivity and vulnerability of offline RL models in unmanned systems.关键词
无人系统/离线强化学习/数据投毒攻击/数据安全Key words
unmanned system/offline reinforcement learning/data poisoning attack/data security分类
信息技术与安全科学引用本文复制引用
周雪,苘大鹏,许晨,吕继光,曾凡一,高朝阳,杨武..无人系统中离线强化学习的隐蔽数据投毒攻击方法[J].通信学报,2024,45(12):16-27,12.基金项目
国家重点研发计划基金资助项目(No.2021YFB3101401) （No.2021YFB3101401）
黑龙江省自然科学基金资助项目(No.TD2022F001) （No.TD2022F001）
国家自然科学基金资助项目(No.U2003206,No.U20B2048,No.U21B2019,No.U22A2036,No.62272127) （No.U2003206,No.U20B2048,No.U21B2019,No.U22A2036,No.62272127）
中央高校基本科研业务费专项资金资助项目(No.3072024XX0607) The National Key Research and Development Program of China(No.2021YFB3101401),The Natural Sci-ence Foundation of Heilongjiang Province(No.TD2022F001),The National Natural Science Foundation of China(No.U2003206,No.U20B2048,No.U21B2019,No.U22A2036,No.62272127),The Fundamental Research Funds of the Central Universities(No.3072024XX0607) （No.3072024XX0607）
