Human-level control through deep reinforcement learning
At a glance
Presented DQN, learning to play 49 Atari games from screen images and game scores.
In one paragraph
Demonstrated the deep Q-network (DQN), a system that learned to play 49 Atari games through trial and error, using only screen images and game scores. Across the games, its performance was comparable to a professional human games tester.
Authors 19
- Volodymyr Mnih
- Koray Kavukcuoglu
- David Silver
- Andrei A. Rusu
- Joel Veness
- Marc G. Bellemare
- Alex Graves
- Martin Riedmiller
- Andreas K. Fidjeland
- Georg Ostrovski
- Stig Petersen
- Charles Beattie
- Amir Sadik
- Ioannis Antonoglou
- Helen King
- Dharshan Kumaran
- Daan Wierstra
- Shane Legg
- Demis Hassabis
