The goal for this project is to create an intelligent poker squares player for the EAAI NSG Challenge as a way to delve into recent research of Monte Carlo Tree Search Methods and other game-based search problems.
You can read a bit about the contest here.
Documentation for the contest code is available here.
The current working implementation involves a genetic programming algorithm that employs evolving decision trees based on rule-based decision types. However, the current implementation does not take advantage of the 30 seconds per game (since decision trees can play the game incredibly quickly).
- Try integrating MCTS implementation for later steps.