Action Elimination and Stopping Conditions for Reinforcement Learning.
Eyal Even-DarShie MannorYishay MansourPublished in: ICML (2003)
Keyphrases
- reinforcement learning
- action selection
- function approximation
- partially observable domains
- sufficient conditions
- action space
- state space
- learning algorithm
- state action
- optimal control
- reward shaping
- optimal policy
- supervised learning
- learning process
- machine learning
- reasoning about actions
- robotic control
- transition model
- learning problems
- reinforcement learning algorithms
- real time
- markov decision processes
- neural network