Deep Reinforcement Learning Agent (DQN) playing Atari Breakout

This video illustrates the improvement in the performance of the DQN agent over training (i.e. after 100, 200, 400 and 600 episodes). After 600 episodes DQN agent finds and exploits the optimal strategy in this game, which is to make a tunnel around the side, and then allow the ball to hit blocks by bouncing behind the wall. Note: the score is displayed at the top left of the screen (maximum for clearing one screen is 448 points), number of lives remaining is shown in the middle (starting with 5 lives), and the “1” on the top right indicates this is a 1-player game.

In the video below you can see all the actions running in the Convolutional Neural Network.

The code was implemented by Nathan Sprague and can be downloaded from here:

https://github.com/spragunr/deep_q_rl

It follows the paper "Playing Atari with Deep Reinforcement Learning" by Volodymyr Mnih, Koray Kavukcuoglu, David Silver, Alex Graves, Ioannis Antonoglou, Daan Wierstra and Martin Riedmiller that you can get from:

http://arxiv.org/abs/1312.5602


— Human-level control through deep reinforcement learning:
https://storage.googleapis.com/deepmind-media/dqn/DQNNaturePaper.pdf
— Open source implementation is available on GitHub:
https://github.com/deepmind/dqn