Nim is an adversarial game in which two players take turns removing objects from four distinct piles.
On each turn, a player must remove some number of objects from a single pile. The goal of the game is to have your
opponent take the last object from the game board.
- AI Algorithm: This is where you can change the AI Algorithm that will play against you. Q-learning is the only algorithm available for Nim at the moment.
- # Training Rounds: If a machine learning algorithm is selected, you will be able to select the number of training rounds the AI plays against itself.
This is how the AI learns how to play the game. The more learning it does, the better it will become.
- Let AI Start: This will let the AI have the first move.
- New Game: This will completely reset the game.
How it works: The AI is ran on the server only.
When you make a move, your move is sent to the server asynchronously, and the AI's move will be delivered back to the player.
All machine learning is done in advance. You are welcome to change training rounds, and even the algorithm, mid-way through a game!