Deep Q-networks for learning control from pixels

What changed

A deep reinforcement-learning system learned action values from visual observations and rewards across a range of Atari games.

Scientific scope

A deep reinforcement-learning system learned action values from visual observations and rewards across a range of Atari games. Scope limitation: This entry credits the human development of a learning method. Performance varies by game; a game benchmark is not evidence of general scientific competence.

Evidence

Assessment uses the catalogue-linked source evidence for the stated contribution. This entry credits the human development of a learning method. Performance varies by game; a game benchmark is not evidence of general scientific competence.

Contributors

• The research collaboration listed in the linked record and source byline · SCIENCE: Collective scoped contribution. The complete recorded byline remains in the research record; this label does not collapse people into one individual.

• Human-level control through deep reinforcement learning
• Human-level control through deep reinforcement learning — deposited bibliography
• DQN original paper and abstract

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction