Work

Strategic Features for General Games

Year: 2021

Type: conference-paper

Abstract: This short paper describes an ongoing research project that requires the automated self-play learning and evaluation of a large number of board games in digital form. We describe the approach we are taking to determine relevant features, for biasing MCTS playouts for arbitrary games played on arbitrary geometries. Benefits of our approach include efficient implementation, the potential to transfer learnt knowledge to new contexts, and the potential to explain strategic knowledge embedded in features in human-comprehensible terms.

Source: Research Publications (Maastricht University)

DOI: 10.48550/arxiv.2101.00843

Authors: Browne, Cameron, Soemers, Dennis J. N. J., Piette, Eric

Institution: Maastricht University

Language: English

Topic: Artificial Intelligence in Games

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W2913687272

FWCI: 0.578

Cites: 16

Cited by: 11

Related to: 10