Will Dabney

E751636

machine learning researcher person

Will Dabney is a machine learning researcher known for his influential work in deep reinforcement learning, including co-developing the Rainbow DQN algorithm.

Try in SPARQL Jump to: Surface forms Statements Referenced by

All labels observed (1)

Label	Occurrences
Will Dabney canonical	1

Statements (41)

Predicate	Object
instanceOf	machine learning researcher ⓘ person ⓘ
citizenship	United States of America ⓘ
coAuthor	Alexandre P. Barreto NERFINISHED ⓘ Georg Ostrovski NERFINISHED ⓘ Hado van Hasselt NERFINISHED ⓘ Marc G. Bellemare NERFINISHED ⓘ Meire Fortunato NERFINISHED ⓘ Nando de Freitas NERFINISHED ⓘ Rémi Munos NERFINISHED ⓘ Shane Legg NERFINISHED ⓘ
doctoralThesisTopic	reinforcement learning ⓘ
educatedAt	Colorado State University NERFINISHED ⓘ University of Massachusetts Amherst NERFINISHED ⓘ
employer	Google DeepMind NERFINISHED ⓘ
fieldOfWork	artificial intelligence ⓘ deep reinforcement learning ⓘ machine learning ⓘ reinforcement learning ⓘ
gender	male ⓘ
hasAcademicAdvisor	Andrew G. Barto NERFINISHED ⓘ
influencedBy	Andrew G. Barto NERFINISHED ⓘ Richard S. Sutton NERFINISHED ⓘ
knownFor	C51 distributional DQN NERFINISHED ⓘ co-developing the Rainbow DQN algorithm ⓘ deep reinforcement learning research ⓘ distributional reinforcement learning ⓘ implicit quantile networks (IQN) ⓘ quantile regression DQN (QR-DQN) ⓘ research on value distribution in reinforcement learning ⓘ
languageSpoken	English ⓘ
memberOf	DeepMind NERFINISHED ⓘ
notableWork	A Distributional Perspective on Reinforcement Learning NERFINISHED ⓘ Distributional Reinforcement Learning with Quantile Regression NERFINISHED ⓘ Implicit Quantile Networks for Distributional Reinforcement Learning NERFINISHED ⓘ Rainbow: Combining Improvements in Deep Reinforcement Learning NERFINISHED ⓘ
researchInterest	distributional value functions ⓘ exploration in reinforcement learning ⓘ representation learning in RL ⓘ value-based reinforcement learning ⓘ
workLocation	London, England ⓘ surface form: London

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.

Rainbow DQN → proposedBy → Will Dabney ⓘ