Reinforcement Learning

By Richard S. Sutton. 1992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

1. Sergey Levine
2. Peter S. Dayan
3. Yoshua Bengio
4. J. Urgen Schmidhuber
5. Nathaniel D. Daw
6. Karl J. Friston
7. Pieter Abbeel
8. John P. O'Doherty

Citations in the corpus

• Reinforcement learning in robotics: A survey (2013, background): “One example is tile coding (Sutton and Barto, 1998), where the space is subdivided into (potentially irregularly shaped) regions, called tiling.”
• Least Squares Temporal Difference Methods: An Analysis under General Conditions (2012, methodology): “, Glynn and Iglehart [GI89], Sutton and Barto [SB98], Precup et al.”
• Multi-Agent Systems by Incremental Gradient Reinforcement Learning (2001, unknown): “We opted for Reinforcement Learning (RL) methods [Sutton and Barto, 1998] as they do not require a teacher knowing before hand a solution of the problem.”
• Inter-module credit assignment in modular reinforcement learning (2003, methodology): “One straightforward method of achieving this condition is first to compute the TD error (Sutton and Barto 1998) for the entire task: δ(t) = r(t) + γV (x(t + 1))− V (x(t)) (6)”