Learning to Play Stochastic Two-player Perfect-Information Games without Knowledge
This work addresses a specific challenge in game AI for stochastic environments, representing an incremental advancement over existing deterministic methods.
The authors tackled the problem of learning and planning in stochastic two-player perfect-information games by extending the Descent framework, achieving the best results with their generalization of Descent compared to state-of-the-art algorithms like Expectiminimax and Polygames.
In this paper, we extend the Descent framework, which enables learning and planning in the context of two-player games with perfect information, to the framework of stochastic games. We propose two ways of doing this, the first way generalizes the search algorithm, i.e. Descent, to stochastic games and the second way approximates stochastic games by deterministic games. We then evaluate them on the game EinStein wurfelt nicht! against state-of-the-art algorithms: Expectiminimax and Polygames (i.e. the Alpha Zero algorithm). It is our generalization of Descent which obtains the best results. The approximation by deterministic games nevertheless obtains good results, presaging that it could give better results in particular contexts.