Learning Atari from pixels

2013

Deep reinforcement learning combines a neural network with Q-learning on Atari games.

A shared learning approach works across multiple games.

Explore on Humanity Unfolds

Sources

• arxiv.org