UserDriven Narrative Variation in Large Story Domains using Monte Carlo Tree Search

13th International Conference on Autonomous Agents and Multiagent Systems

Bilal Kartal ; John Koenig ; Stephen J. Guy

PID  :  119279

来源: CEUR

【 摘 要 】

Planningbased techniques are powerful tools for automated narrative generation, however, as the planning domain grows in the number of possible actions traditional planning tech niques suffer from a combinatorial explosion. In this work, we apply Monte Carlo Tree Search to goaldriven narrative generation. We demonstrate our approach to have an or der of magnitude improvement in performance over tradi tional search techniques when planning over large story do mains. Additionally, we propose a Bayesian story evaluation method to guide the planning towards believable narratives which achieve userdefined goals. Finally, we present an in teractive user interface which enables users of our framework to modify the believability of different actions, resulting in