RL agent for pong

Repository containing the final project of the reinforcement learning course of the EIT Digital data science master at Aalto University 📚.

Objective

The objective of this project is to train a reinforcement learning agent which is able to play pong game. We have used the Pong Gym environment.

RL algorithms selected

For the training of the agent we have used the PPO algorithm, which perform comparably or better than state-of-the-art approaches while being much simpler to implement and tune.

Authors

• Cristian Abrante - CristianAbrante
• Alejandro Esquivias - ESQ0001