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 2012.12186cs.AI
收藏

通过模仿Oracle规划师来学习玩不完全信息游戏

Learning to Play Imperfect-Information Games by Imitating an Oracle Planner

Aalto University(阿尔托大学);Supercell

Rinu Boney, Alexander Ilin, Juho Kannala, Jarno Seppänen

摘要:

我们考虑学习玩多人同时移动和大的状态动作空间的信息博弈。以前解决这种具有挑战性的游戏的尝试主要集中在无模型学习方法上,通常需要数百年的经验才能产生有竞争力的代理。我们的方法是基于模型的规划。我们首先建立一个(预言机)计划,可以访问的全部状态的环境,然后提取的知识的预言机(追随者)代理,这是训练发挥的信息博弈,通过模仿预言机的选择,解决部分可观测性的问题。我们的实验表明,规划与天真的蒙特卡罗树搜索并没有很好地执行大型组合动作空间。因此,我们提出了一个固定深度的树搜索和解耦汤普森采样的行动选择规划。我们表明,规划者能够发现有效的播放策略,在游戏中的《皇室战争》和Pommerman和追随者的政策成功地学会了执行他们的培训几百场战斗。

英文摘要:

We consider learning to play multiplayer imperfect-information games with simultaneous moves and large state-action spaces. Previous attempts to tackle such challenging games have largely focused on model-free learning methods, often requiring hundreds of years of experience to produce competitive agents. Our approach is based on model-based planning. We tackle the problem of partial observability by first building an (oracle) planner that has access to the full state of the environment and then distilling the knowledge of the oracle to a (follower) agent which is trained to play the imperfect-information game by imitating the oracle's choices. We experimentally show that planning with naive Monte Carlo tree search does not perform very well in large combinatorial action spaces. We therefore propose planning with a fixed-depth tree search and decoupled Thompson sampling for action selection. We show that the planner is able to discover efficient playing strategies in the games of Clash Royale and Pommerman and the follower policy successfully learns to implement them by training on a few hundred battles.

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。

↑