STRATEGA: A General Strategy Games Framework-猫眼课题宝

DOI：

发表时间：

2020

期刊：

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Alexander Dockhorn;Jorge Hurtado Grueso;Dominik Jeurissen;Diego Perez Liebana

战略游戏是一种复杂的环境，经常用于人工智能研究以评估新的算法。尽管大多数策略游戏具有共性，但研究往往只专注于一种游戏，这可能会导致偏见或过度ﬁ设置到特定的环境。在本文中，我们激励并提出了S TRATEGA-一个通用的策略游戏框架，用于玩基于回合的n人实时策略游戏。该平台目前实现了基于回合的游戏，可以通过Yaml-ﬁLes欺骗ﬁ。它公开了一个可以访问远期模型的API，以促进对统计远期规划代理的研究。该框架和代理可以在游戏过程中记录信息，用于分析和调试算法。我们还给出了一些基于规则的代理，以及蒙特卡罗树搜索和滚动地平线进化等基于搜索的代理，并对它们的性能进行了定量的分析，以演示框架的使用。结果，虽然纯粹是说明性的，但显示了传统的基于搜索的代理在处理这些游戏中的高分支因素时存在的已知问题。S特技表演可在以下网址下载：https://github.com/GAIGResearch/Stratega

Strategy games are complex environments often used in AI-research to evaluate new algorithms. Despite the common-alities of most strategy games, often research is focused on one game only, which may lead to bias or overﬁtting to a particular environment. In this paper, we motivate and present S TRATEGA - a general strategy games framework for playing n-player turn-based and real-time strategy games. The platform currently implements turn-based games, which can be conﬁgured via YAML-ﬁles. It exposes an API with access to a forward model to facilitate research on statistical forward planning agents. The framework and agents can log information during games for analysing and debugging algorithms. We also present some sample rule-based agents, as well as search-based agents like Monte Carlo Tree Search and Rolling Horizon Evolution, and quantitatively analyse their performance to demonstrate the use of the framework. Results, although purely illustrative, show the known problems that traditional search-based agents have when dealing with high branching factors in these games. S TRATEGA can be downloaded at: https://github.com/GAIGResearch/Stratega

STRATEGA: A General Strategy Games Framework

文献摘要

相似文献