https://github.com/githubjakob/mcts-tictactoe-connectfour
Two Games (Tic Tac Toe and Connect Four) with a MCTS based AI opponent
https://github.com/githubjakob/mcts-tictactoe-connectfour
Last synced: about 1 year ago
JSON representation
Two Games (Tic Tac Toe and Connect Four) with a MCTS based AI opponent
- Host: GitHub
- URL: https://github.com/githubjakob/mcts-tictactoe-connectfour
- Owner: githubjakob
- Created: 2018-04-19T10:18:41.000Z (about 8 years ago)
- Default Branch: master
- Last Pushed: 2018-05-18T07:30:05.000Z (about 8 years ago)
- Last Synced: 2025-02-16T20:28:22.652Z (over 1 year ago)
- Language: Java
- Homepage:
- Size: 22.5 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# MCTS based AI opponent for TicTacToe and Connect Four
You can play the games in the terminal by running the main method. Both games take the player's input as
xy-coordinates (with the top left corner as 00). The AI opponent will then select the most promising move
based on a search tree built using Monte Carlo tree search. Specifically the MCTS implementation uses UCT in the
selection step of the algorithm.
## Credit
The implementation of the Monte Carlo Tree Search was build with help of this tutorial:
http://www.baeldung.com/java-monte-carlo-tree-search