My attempt to run DQN on Jetson TX1, which learns how to play Nintendo Famicom Mini games through reinforcement learning directly. - View it on GitHub