Loading timeline…
19891980s
Christopher Watkins
Computer scientist who invented Q-learning, a cornerstone of Reinforcement Learning.
Organizations
Cambridge UniversityUniversity of London
Major Achievements
- •Formulated Q-learning in his 1989 PhD dissertation 'Learning from Delayed Rewards'.
- •Provided the first mathematical proof that an agent could learn an optimal policy without requiring an environmental model.
- •His algorithm was eventually combined with deep neural networks by DeepMind in 2013 to master Atari games.
Key Papers
- Q-Learning
Christopher J. C. H. Watkins, Peter Dayan · Machine Learning · 1992
Videos
Friends, rivals and ‘face-offs’ | Christopher Watkins | TEDxAbertayUniversity
TEDx Talks