https://github.com/sykwer/pg_pong

Train ATARI pong agent by stochastic policy gradient method from raw playing images.
https://github.com/sykwer/pg_pong

neural-network policy-gradient python3 reinforcement-learning

Last synced: 2 months ago
JSON representation

Train ATARI pong agent by stochastic policy gradient method from raw playing images.

• Host: GitHub
• URL: https://github.com/sykwer/pg_pong
• Owner: sykwer
• Created: 2018-03-02T11:42:58.000Z (over 8 years ago)
• Default Branch: master
• Last Pushed: 2018-03-08T08:49:27.000Z (over 8 years ago)
• Last Synced: 2026-01-28T09:57:35.204Z (6 months ago)
• Topics: neural-network, policy-gradient, python3, reinforcement-learning
• Language: Python
• Homepage:
• Size: 9.31 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# pg_pong

## Policy Gradient Method (powered by NeruralNet)

Implementation of Policy Gradient method that uses NeuralNet for policy function approximation.

This implementation does not depend on any ML library.

## How to use

```

git clone https://github.com/sykwer/pg_pong.git

cd pg_pong

python main.py

```

## Duaring training

![demo](https://github.com/sykwer/pg_pong/blob/master/movie/atari_pong_movie.mov.gif)