Skip to content

riouh/kaigi_reinforcement_learning

Repository files navigation

Reinforcement learning introduction

This repo proposes a small UI to train an test Q-Learning agents on a simple game. The UI is made with Svelte and Typescript. The Q-learning algorithm runs in a nodeJS server and the communication with the UI is made via Websocket to give a real-time sensation to the user during the learning phase.

You can use it in local. For that, you will need to have nodeJS 16 and npm 8 installed.

cd rl-demo npm install npm run dev

Then, access the UI on http://localhost:3002

About

Easy game to demonstrate how Q-Learning works.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published