Decomposing Drama Management in Educational Interactive Narrative: A Modular Reinforcement Learning Approach
Decomposing Drama Management in Educational Interactive Narrative: A Modular Reinforcement Learning Approach
复制标题
教育互动叙事中的分解戏剧管理：模块化强化学习方法
DOI：
10.1007/978-3-319-48279-8_24
复制
发表时间：
2016
期刊：
影响因子：
4.2
通讯作者：
James C. Lester
中科院分区：
文献类型：
--
作者：
Pengcheng Wang;Jonathan P. Rowe;Bradford W. Mott;James C. Lester
近年来，人们对数据驱动的个性化交互式叙事生成和戏剧管理方法越来越感兴趣。强化学习（RL）显示出基于玩家交互数据语料库动态塑造交互式叙事的训练策略的特别前景。一个重要的开放问题是如何设计基于强化学习的戏剧管理器，以便有效地利用玩家交互数据，这些数据通常收集起来很昂贵，并且相对于戏剧管理所需的庞大状态和动作空间来说很稀疏。我们研究了一个离线优化框架，用于在教育互动叙事中培训基于模块化强化学习的戏剧经理，水晶岛。我们利用重要性抽样评估戏剧经理的政策来自不同的分解表示的互动叙事。实证结果表明，戏剧经理质量显着改善，采用优化的模块化RL分解相比，竞争表示。
Recent years have seen growing interest in data-driven approaches to personalized interactive narrative generation and drama management. Reinforcement learning (RL) shows particular promise for training policies to dynamically shape interactive narratives based on corpora of player-interaction data. An important open question is how to design reinforcement learning-based drama managers in order to make effective use of player interaction data, which is often expensive to gather and sparse relative to the vast state and action spaces required by drama management. We investigate an offline optimization framework for training modular reinforcement learning-based drama managers in an educational interactive narrative, Crystal Island. We leverage importance sampling to evaluate drama manager policies derived from different decompositional representations of the interactive narrative. Empirical results show significant improvements in drama manager quality from adopting an optimized modular RL decomposition compared to competing representations.