A near-optimal closed-hand Chinese Poker AI
Switch branches/tags
Nothing to show
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
strategy
train
LICENSE.txt
README.md
coefficients.data
cpoker.go
evaluator.go
play_test.go

README.md

Near-optimal closed-hand Chinese Poker Player

This repository contains code that can train and evaluate an AI player for the game of closed-hand Chinese Poker.

A description of the game and a summary of how this code works can be found in this article.

Coefficients for a pre-trained player can be found in coefficients.data.

The code is MIT licensed, and you can read the details in LICENSE.txt