Skip to content

Reinforcement Learning: Monte Carlo Search Tree Agent

Notifications You must be signed in to change notification settings

bgyarbro/chopsticks-game

Repository files navigation

chopsticks

Reinforcement Learning

To try out the game, go here

https://colab.research.google.com/github/bgyarbro/chopsticks/blob/master/chopsticks_colab.ipynb

Chopsticks is a simple combinatorial. This implementation is in Python. Using the command line interface, User can play against a monte carlo search tree algorithm, or have the computer play one algorithm against another, with the option for random move choice.

Better documentation to come.

Here are the rules to the game. https://en.wikipedia.org/wiki/Chopsticks_(hand_game)

Hand representation. Player 1 [left hand, right hand] Player 2 [left hand, right hand]

GUI in development.

The code for the game was adapted from the tic-tac-toe example in this book. Book here.

starting with the most simple ruleset, then working up to analyzing stephen's split and play in one turn

coded from the ground up in python the initial prototype was in c++

About

Reinforcement Learning: Monte Carlo Search Tree Agent

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published