在黑暗中交友:部分可观测性下的Ad Hoc团队协作
Making Friends in the Dark: Ad Hoc Teamwork Under Partial Observability
AI总结:
该研究定义了部分可观测下的ad hoc团队协作场景,提出仅靠先验知识和部分观测的模型驱动方法,在70个POMDP上验证了其协助未知队友的有效性与扩展性。
摘要:
本文对部分可观测性下的ad hoc团队协作场景给出了正式定义,并提出了一种基于第一性原理的模型驱动方法,该方法仅依赖先验知识与对环境的部分观测即可开展ad hoc团队协作。我们提出了三项区别于以往研究的独特假设:i)环境状态始终是部分可观测的;ii)ad hoc智能体始终无法获取队友的动作信息;iii)ad hoc智能体无法获取可用于从零学习任务的奖励信号。我们在来自11个领域的70个POMDP(部分可观测马尔可夫决策过程)上进行了实验,结果表明该方法不仅能有效协助未知队友解决未知任务,而且在扩展到更具挑战性的问题时也具备鲁棒性。
英文摘要:
This paper introduces a formal definition of the setting of ad hoc teamwork under partial observability and proposes a first-principled model-based approach which relies only on prior knowledge and partial observations of the environment in order to perform ad hoc teamwork. We make three distinct assumptions that set it apart previous works, namely: i) the state of the environment is always partially observable, ii) the actions of the teammates are always unavailable to the ad hoc agent and iii) the ad hoc agent has no access to a reward signal which could be used to learn the task from scratch. Our results in 70 POMDPs from 11 domains show that our approach is not only effective in assisting unknown teammates in solving unknown tasks but is also robust in scaling to more challenging problems.
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。