arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.05528cs.AIcs.MAcs.RO
收藏

对抗性多智能体系统中的合作行为学习

Learning cooperative behaviours in adversarial multi-agent systems

Paris-Saclay University(巴黎萨克雷大学);University of Lincoln(林肯大学);University of York(约克大学)

Ni Wang, Gautham P. Das, Alan G. Millard

摘要:

这项工作扩展了现有的虚拟多智能体平台RoboSumo,以创建TripleSumo -一个用于研究多智能体在连续动作空间中的合作行为的平台,在对抗环境中进行物理接触。在本文中,我们调查的情况下,两个代理,即“错误”和“蚂蚁”,必须组队,并推动另一个代理“蜘蛛”的舞台。为了实现这一目标,新添加的代理“Bug”在“Ant”和“Spider”之间进行的比赛中进行训练。“Bug”必须发展对其他代理的行动的意识,推断双方的策略,并最终学习合作的行动策略。强化学习算法深度确定性策略梯度(DDPG)是通过结合密集和稀疏奖励的混合奖励结构来实现的。通过平均获胜概率和平均获胜所需的步骤数对合作行为进行定量评价。

英文摘要:

This work extends an existing virtual multi-agent platform called RoboSumo to create TripleSumo -- a platform for investigating multi-agent cooperative behaviors in continuous action spaces, with physical contact in an adversarial environment. In this paper we investigate a scenario in which two agents, namely `Bug' and `Ant', must team up and push another agent `Spider' out of the arena. To tackle this goal, the newly added agent `Bug' is trained during an ongoing match between `Ant' and `Spider'. `Bug' must develop awareness of the other agents' actions, infer the strategy of both sides, and eventually learn an action policy to cooperate. The reinforcement learning algorithm Deep Deterministic Policy Gradient (DDPG) is implemented with a hybrid reward structure combining dense and sparse rewards. The cooperative behavior is quantitatively evaluated by the mean probability of winning the match and mean number of steps needed to win.

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。

↑