Loading timeline…
19891980s
Why It Was Important
Introduced by Christopher Watkins, Q-learning allows an agent to learn an optimal action-selection policy for any given computational state without requiring a model of the environment. It became central to modern Reinforcement Learning, eventually powering systems that play Atari, Go, and navigate robots.
Who Invented It
Christopher Watkins
British computer scientist who formulated Q-learning in his PhD thesis.
Applications
- Robotics Navigation
- Game Playing
- Traffic Control Systems
Key Papers
- Q-Learning
Christopher J. C. H. Watkins, Peter Dayan · Machine Learning · 1992
Videos
Reinforcement Learning - Computerphile
Computerphile