https://github.com/honzabrecka/q-p
Yet another implementation of tic-tac-toe game using Q-learning.
https://github.com/honzabrecka/q-p
clojurescript machine-learning q-learning tictactoe
Last synced: 2 months ago
JSON representation
Yet another implementation of tic-tac-toe game using Q-learning.
- Host: GitHub
- URL: https://github.com/honzabrecka/q-p
- Owner: honzabrecka
- License: mit
- Created: 2016-08-04T20:56:57.000Z (almost 10 years ago)
- Default Branch: master
- Last Pushed: 2017-08-08T21:03:58.000Z (almost 9 years ago)
- Last Synced: 2025-01-17T01:41:34.748Z (over 1 year ago)
- Topics: clojurescript, machine-learning, q-learning, tictactoe
- Language: Clojure
- Homepage: https://honzabrecka.com/q-p/
- Size: 153 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# q-p
Yet another implementation of tic-tac-toe game using Q-learning.
## Usage
```cljs
(let [{:keys [x o]} (train q-player q-player 200000)]
(play x o))
```
## Next steps
Implement neural network to be able to handle larger grids & more complex states generally.