Action Refinement in Reinforcement Learning by Probability Smoothing.
Thomas G. DietterichDídac BusquetsRamón López de MántarasCarles SierraPublished in: ICML (2002)
Keyphrases
- reinforcement learning
- action selection
- partially observable domains
- action space
- function approximation
- probability distribution
- markov decision processes
- reward shaping
- transition model
- optimal policy
- temporal difference
- human actions
- learning process
- reinforcement learning algorithms
- probability theory
- multi agent
- supervised learning
- state space
- optimal control
- learning problems
- partially observable
- probabilistic model
- temporal difference learning
- smoothing methods
- multi agent reinforcement learning
- dynamic programming
- video sequences
- robotic control
- learning algorithm