紀要論文

DQNと生物学的制約を用いた人間らしい振る舞いのゲームAI

Video-Game Agents with Human-Like Behavior Using DQN and Biological Constraints

jpn

24321192

https://doi.org/10.15002/00026291

法政大学大学院情報科学研究科

法政大学大学院紀要. 情報科学研究科編

18

1

6

2023-03-24

Version of Record

Video game agents that surpass humans always select the optimal behavior, which may make them look mechanical and uninteresting to human players and audience. Since score-oriented game agents have been almost achieved, a next goal should be to entertain human players and audience by realizing agents that reproduce human-like behavior. A previous method implemented such game agents by introducing biological constraints into Q-learning and A^* search. In this paper, we propose video game agents with more entertaining and more practical human-like behavior by applying biological constraints into the deep Q-network (DQN). Especially, to reduce the problem of the conspicuous mechanical behavior found in the previous method, we improve the method of the biological constraint “tired”, and propose additional biological constraints “confusion” and “carelessness”. We implemented our method in the video game “Infinite Mario Bros.” and conducted two types of experiments that were subjective evaluation. One is to evaluate the behavior of game agents that improved “tired”. The other is to evaluate the behavior of game agents that introduced “confusion” and “carelessness” The results of both experiments indicated that the agents implemented with our method were rated more human-like than those implemented with the previous method.

Article

AA12746425