Curiosity-Driven and Victim-Aware Adversarial Policies
文献类型：会议论文
| 作者 | Gong C(龚晨)3 ; Yang Z(杨洲)1; Bai YP(白云鹏)3; Shi JK(史杰克)1; Sinha Arunesh2; Xu BW(徐博文)1; Lo David1; Hou XW(侯新文)3; Fan GL(范国梁)3
|
| 出版日期 | 2023-05 |
| 会议日期 | December 5-9, 2022 |
| 会议地点 | Austin TX, USA |
| 英文摘要 | Recent years have witnessed great potential in applying Deep Reinforcement Learning (DRL) in various challenging applications, such as autonomous driving, nuclear fusion control, complex game playing, etc. However, recently researchers have revealed that deep reinforcement learning models are vulnerable to adversarial attacks: malicious attackers can train adversarial policies to tamper with the observations of a well-trained victim agent, the latter of which fails dramatically when faced with such an attack. Understanding and improving the adversarial robustness of deep reinforcement learning is of great importance in enhancing the quality and reliability of a wide range of DRL-enabled systems.
In this paper, we develop curiosity-driven and victim-aware adversarial policy training, a novel method that can more effectively exploit the defects of victim agents. To be victim-aware, we build a surrogate network that can approximate the state-value function of a black-box victim to collect the victim’s information. Then we propose a curiosity-driven approach, which encourages an adversarial policy to utilize the information from the hidden layer of
the surrogate network to exploit the vulnerability of victims efficiently. Extensive experiments demonstrate that our proposed method outperforms or achieves a similar level of performance as the current state-of-the-art across multiple environments. We perform an ablation study to emphasize the benefits of utilizing the approximated victim information. Further analysis suggests that
our method is harder to defend against a commonly used defensive strategy, which calls attention to more effective protection on the systems using DRL.
|
| 源URL | [http://ir.ia.ac.cn/handle/173211/52195]  |
| 专题 | 自动化研究所_复杂系统管理与控制国家重点实验室_机器人应用与理论组 |
| 作者单位 | 1.Singapore Management University 2.Rutgers University 3.Institute of Automation, Chinese Academy of Sciences, China |
| 推荐引用方式 GB/T 7714 | Gong C,Yang Z,Bai YP,et al. Curiosity-Driven and Victim-Aware Adversarial Policies[C]. 见:. Austin TX, USA. December 5-9, 2022. |
入库方式： OAI收割
来源：自动化研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。