TY - JOUR
T1 - ATS-O2A
T2 - A state-based adversarial attack strategy on deep reinforcement learning
AU - Li, Xiangjuan
AU - Li, Yang
AU - Feng, Zhaowen
AU - Wang, Zhaoxuan
AU - Pan, Quan
N1 - Publisher Copyright:
© 2023 Elsevier Ltd
PY - 2023/6
Y1 - 2023/6
N2 - In recent years, deep reinforcement learning has been widely applied in many decision-making tasks requiring high safety and security due to its excellent performance. However, if an adversary attacks when the agent making critical decisions, it is bound to bring disastrous consequences because humans cannot detect it. Therefore, it is necessary to study adversarial attacks against deep reinforcement learning to help researchers design highly robust and secure algorithms and systems. In this paper, we proposed an attack method based on Attack Time Selection (ATS) function and Optimal Attack Action (O2A) strategy, named ATS-O2A. We select the critical attack moment through the ATS function, and then combine the state-based strategy with the O2A strategy to select the optimal attack action which has profound influence as targeted action, finally we launch an attack by making targeted adversarial examples. In order to measure the stealthiness and effectiveness of the attack, we designed a new measurement index. Experiments show that our method can effectively reduce unnecessary attacks and improve the efficiency of attacks.
AB - In recent years, deep reinforcement learning has been widely applied in many decision-making tasks requiring high safety and security due to its excellent performance. However, if an adversary attacks when the agent making critical decisions, it is bound to bring disastrous consequences because humans cannot detect it. Therefore, it is necessary to study adversarial attacks against deep reinforcement learning to help researchers design highly robust and secure algorithms and systems. In this paper, we proposed an attack method based on Attack Time Selection (ATS) function and Optimal Attack Action (O2A) strategy, named ATS-O2A. We select the critical attack moment through the ATS function, and then combine the state-based strategy with the O2A strategy to select the optimal attack action which has profound influence as targeted action, finally we launch an attack by making targeted adversarial examples. In order to measure the stealthiness and effectiveness of the attack, we designed a new measurement index. Experiments show that our method can effectively reduce unnecessary attacks and improve the efficiency of attacks.
KW - Adversarial attack
KW - Deep learning security
KW - Deep reinforcement learning
KW - Machine learning
KW - Targeted attack
UR - http://www.scopus.com/inward/record.url?scp=85152226426&partnerID=8YFLogxK
U2 - 10.1016/j.cose.2023.103259
DO - 10.1016/j.cose.2023.103259
M3 - 文章
AN - SCOPUS:85152226426
SN - 0167-4048
VL - 129
JO - Computers and Security
JF - Computers and Security
M1 - 103259
ER -