/pong_rl

Miniproject | Learning to play Pong with Deep Reinforcement Learning (Artificial Neural Network 2020)

Primary LanguageJupyter Notebook

Pong

Miniproject | Learning to play Pong with Deep Reinforcement Learning

Pong is a game environment provided by OpenAI. The purpose of the project was to implement a model able to beat an artificial adversary at the game of Pong.

We did it by implementing the gradient-based RL algorithm commonly known as REINFORCE in a neural network. Refer to the notebook pong_rl.ipynb for more information and the code of our implementation.