arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2205.09362cs.AI
收藏

多智能体强化学习中的稀疏对抗攻击

Sparse Adversarial Attack in Multi-agent Reinforcement Learning

Yizheng Hu, Zhihua Zhang

摘要:

协作多智能体强化学习(cMARL)有很多实际应用,但现有cMARL算法训练的策略在部署时不够鲁棒。针对强化学习系统的对抗攻击也有很多方法,这意味着强化学习系统可以遭受对抗攻击,但大多数研究都集中在单Agent强化学习上。在本文中,我们提出了一个\textit{sparse adversarial attack}的cMARL系统。我们使用带有正则化的(MA)RL来训练攻击策略。我们的实验表明,当前cMARL算法训练的策略在团队中只有一个或几个代理时(例如,8个中的1个或25个中的5个)在几个时间步(例如,总共40个时间步中的攻击3)。

英文摘要:

Cooperative multi-agent reinforcement learning (cMARL) has many real applications, but the policy trained by existing cMARL algorithms is not robust enough when deployed. There exist also many methods about adversarial attacks on the RL system, which implies that the RL system can suffer from adversarial attacks, but most of them focused on single agent RL. In this paper, we propose a \textit{sparse adversarial attack} on cMARL systems. We use (MA)RL with regularization to train the attack policy. Our experiments show that the policy trained by the current cMARL algorithm can obtain poor performance when only one or a few agents in the team (e.g., 1 of 8 or 5 of 25) were attacked at a few timesteps (e.g., attack 3 of total 40 timesteps).

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。

↑