A Study of Reinforcement Learning in the Continuous Case by the Means of Viscosity Solutions

By R´emi Munos. 2000

90 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics, Management Science and Operations Research

How researchers see this paper

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

In Artificial Intelligence

1. Shie Mannor
2. Csaba Szepesvari
3. R´emi Munos
4. Sergey Levine
5. Lihong Li
6. Yaakov Engel
7. Marc G. Bellemare
8. Andrew W. Moore

In Computational Theory and Mathematics

1. Andrew W. Moore
2. Shie Mannor
3. R. Meir
4. Yaakov Engel
5. Dongbing Gu
6. Erfu Yang
7. R´emi Munos
8. Sergey Levine

In Management Science and Operations Research

1. Shie Mannor
2. R´emi Munos
3. Andrew W. Moore
4. R. Meir
5. Yaakov Engel
6. Dongbing Gu
7. Erfu Yang
8. Csaba Szepesvari

Citations in the corpus

• Bayes Meets Bellman: The Gaussian Process Approach to Temporal Difference Learning (2003, methodology): “…fuzzy sets, hard and soft state aggregation, and non-parametric methods such as variable resolution, instance-based learning , viscosity solutions (Munos, 2000) and local averaging (Ormoneit & Sen, 2002) (We refer the reader Sutton and Barto (1998) for a review of all but the last two of these…”
• Variable Resolution Discretization in Optimal Control (2002, methodology): “We refer to Kushner and Dupuis (1992) for the process of discretizing a continuous time and space optimal control problem into a finite Markov Decision Process (MDP), and to Munos (2000) for similar methods in reinforcement learning.”
• Multiagent Reinforcement Learning for Multi-Robot Systems: A Survey (2004, background): “Based on the theoretic framework of viscosity solutions, Munos [74] conducted a study of RL for the continuous state-space and time control problems.”