2019/03/21 by Dennis J. N. J. Soemers, Éric Piette, Soemers, Dennis J. N. J. +3
Computer Science · Economics, Econometrics and Finance · Mathematics · Psychology · Social Sciences · #Artificial Intelligence (cs.AI) #Artificial Intelligence in Games #Digital Games and Media #Educational Games and Gamification #FOS: Computer and information sciences #Machine Learning (stat.ML) #Sports Analytics and Performance #cs.AI #stat.ML
paper · pdf · doi:10.48550/arxiv.1903.08942
Accepted at IEEE CEC 2019, Special Session on Games. Copyright of final version held by IEEE
arxiv created 2019/03/21 · openalex publication_date 2019/03/21 · arxiv updated 2019/03/22 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
This paper proposes using a linear function approximator, rather than a deep neural network (DNN), to bias a Monte Carlo tree search (MCTS) player for general games. This is unlikely to match the potential raw playing strength of DNNs, but has advantages in terms of generality, interpretability and resources (time and hardware) required for training. Features describing local patterns are used as inputs. The features are formulated in such a way that they are easily interpretable and applicable to a wide range of general games, and might encode simple local strategies. We gradually create new features during the same self-play training process used to learn feature weights. We evaluate the playing strength of an MCTS player biased by learnt features against a standard upper confidence bounds for trees (UCT) player in multiple different board games, and demonstrate significantly improved playing strength in the majority of them after a small number of self-play training games.