Transformer Based Planning in the Observation Space with Applications to Trick Taking Card Games

arXiv:2404.13150v1 Announce Type: cross
Abstract: Traditional search algorithms have issues when applied to games of imperfect information where the number of possible underlying states and trajectories are very large. This challenge is particularly evident in trick-taking card games. While state sampling techniques such as Perfect Information Monte Carlo (PIMC) search has shown success in these contexts, they still have major limitations.
We present Generative Observation Monte Carlo Tree Search (GO-MCTS), which utilizes MCTS on observation sequences generated by a game …

abstract algorithms applications arxiv card challenge cs.ai cs.lg games information observation planning sampling search search algorithms space state transformer trick type