Publication
GameTable Working Group 1 Meeting Report on Search, Planning, Learning, and Explainability
Authors:
Dennis J. N. J. Soemers, Jakub Kowalski, Eric Piette, Achille Morenville, Walter Crist
Venue:
International Computer Games Association (ICGA) Journal, 2024
Topics:
GameTable, general game playing, reinforcement learning, explainability, imperfect information, human-like AI
Links: PDF · ICGA Journal
Abstract
This paper reports on the first in-person meeting of Working Group 1 of the GameTable COST Action, focused on search, planning, learning, and explainability in game AI.
The meeting brought together researchers from multiple disciplines to discuss three key research directions: human-like game-playing AI, general approaches for imperfect-information games, and explainable search and reinforcement learning methods.
The report summarises presentations, discussions, and emerging challenges, and outlines promising directions for future collaborative research within the GameTable network.
Context
This work is part of the GameTable COST Action, a European research network exploring tabletop games through artificial intelligence, cultural heritage, and computational methods.
The Working Group focuses on advancing general game playing AI, with particular attention to:
- Developing human-like AI players
- Handling imperfect-information games in general frameworks
- Designing explainable AI methods for games
These directions extend ongoing work from the Digital Ludeme Project and aim to better understand how humans play games across cultures and time.
Full reference
Soemers, D. J. N. J., Kowalski, J., Piette, E., Morenville, A., Crist, W. (2024). GameTable Working Group 1 Meeting Report on Search, Planning, Learning, and Explainability. ICGA Journal, 46, 28–35.
BibTeX
@article{soemers2024gametable_wg1,
author = {Soemers, Dennis J. N. J. and Kowalski, Jakub and Piette, Eric and Morenville, Achille and Crist, Walter},
title = {GameTable Working Group 1 Meeting Report on Search, Planning, Learning, and Explainability},
journal = {ICGA Journal},
volume = {46},
pages = {28--35},
year = {2024},
doi = {10.3233/ICG-240251}
}