(A)I Can Play Gomoku: An Intelligent Tutoring System for Strategic Games
Companion Proceedings of the Annual Symposium on Computer-Human Interaction in Play
2025
Abstract
In this demo, we present a Gomoku Tutor that can provide users with immediate feedback and hints, and support users in engaging in after-action reviews when playing the strategy board game Gomoku. These capabilities are powered by an expert model that we discovered using AlphaZero. Our work fills the gap of developing and deploying tutors in domains where there are multiple approaches to tackle a problem. Furthermore, we also demonstrated the feasibility of using black-box models such as AlphaZero to improve human learning, reasoning, and strategic decision-making. Our work aims to introduce new pedagogical uses for Game AI models and contribute to the broader scope of intelligent tutoring systems (ITS).
BibTeX
@inproceedings{zhang-chiplay-2025,
title = {(A)I Can Play Gomoku: An Intelligent Tutoring System for Strategic Games},
author = {Zhang, Qiao and Wang, Chunyi and MacLellan, Christopher J.},
booktitle = {Companion Proceedings of the Annual Symposium on Computer-Human Interaction in Play},
pages = {239-244},
year = {2025},
doi = {10.1145/3744736.3749191},
}