Automatic Feature Selection for Model-Based Reinforcement Learning in Factored MDPs.
Mark KroonShimon WhitesonPublished in: ICMLA (2009)
Keyphrases
- model based reinforcement learning
- factored mdps
- markov decision processes
- markov decision problems
- policy iteration
- state space
- reinforcement learning
- optimal policy
- finite state
- dynamic programming
- planning under uncertainty
- partially observable
- decision processes
- decision theoretic
- infinite horizon
- average cost
- markov decision process
- learning algorithm