“Deep Q-Network, a neural network that could play Atari games at a superhuman level, used as a project goal for learning AI.”

“reproduce the DQN results”

“Deep Q-Network, the algorithm developed by DeepMind that trained itself from scratch to play Atari games at a superhuman level.”

“the standard Atari games the way since DQN 12 years ago or whatever have been tested in this regime where you take one agent, you train it for 200 million frames”

“The first end-to-end learning system that could master Atari games from raw pixels.”