extend Policy to allow for uniform random (current) and MCTS

keithgw / wingspan

A Wingspan AI that uses reinforcement learning to learn to play the wingspan board game.

2 stars 1 forks source link

extend Policy to allow for uniform random (current) and MCTS #48

Closed keithgw closed 9 months ago

keithgw commented 9 months ago

When BotPlayer is initiated, it can either follow a uniform random policy, which is the only policy currently implemented, or the new MCTS policy. Later MCTS policy can be parameterized by its reward type and explore/exploit strategy.