TY - JOUR AU - Yang, Zuozhi AU - Ontañón, Santiago PY - 2019/10/08 Y2 - 2024/03/28 TI - Guiding Monte Carlo Tree Search by Scripts in Real-Time Strategy Games JF - Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment JA - AIIDE VL - 15 IS - 1 SE - Full Oral Papers DO - 10.1609/aiide.v15i1.5231 UR - https://ojs.aaai.org/index.php/AIIDE/article/view/5231 SP - 100-106 AB - <p>In Real-Time Strategy (RTS) games, the action space grows combinatorially with respect to the number of units. With limited computing budget between actions, methods like Monte Carlo Tree Search (MCTS) tend to get lost in the massive search space. An interesting line of existing work is to incorporate human knowledge in the form of scripts. In this paper, we investigate different possibilities for incorporating scripts into the tree policy while still maintaining the convergence guarantees of MCTS. We also report experiments on incorporating the scripts into the playout policy, which showed that unbiased bots perform better than biased bots.</p> ER -