The CTB.py file contains algorithm for an Q-Learning agent to learn to catch the ball. The GUI interface is written using the Pygame gaming environment.

The ppo.py file is implementation of proximal policy gradient.

The code for pong_policy_gradient.py is taken from Andrej Karpathy's 'Pong from Pixels' blog.

The tf_pg_pong.py file is an implementation of pong_policy_gradient.py in Tensorflow.

The ddpg.py file is implementation of deep deterministic policy gradient.