Crate arbor[][src]

This library provides a generic interface to the Monte Carlo Tree Search (MCTS) algorithm. It can be used as a general game playing agent for two-player board games.

Structs

MCTS

This struct provides methods to set search parameters and control execution. It uses a builder pattern allowing only the desired parameters to be changed from default.

Enums

GameResult

This enum describes the result of a game.

Traits

Action

This trait describes an allowed move for a game state. This type is passed to the “make” function to produce the next game state. The algorithm keeps track of all allowed actions for each game state that is visited. Limit the size of this type and prefer a contiguous memory layout for best performance (e.g. enum, integer).

GameState

This trait describes the current state of the game from which to begin searching for the best move.