The value functions of Markov decision processes.
Ehud LehrerEilon SolanOmri N. SolanPublished in: Oper. Res. Lett. (2016)
Keyphrases
- markov decision processes
- reinforcement learning
- optimal policy
- dynamic programming
- state space
- finite state
- reinforcement learning algorithms
- reachability analysis
- transition matrices
- decision theoretic planning
- decision diagrams
- policy iteration
- factored mdps
- model based reinforcement learning
- markov decision process
- finite horizon
- decision processes
- risk sensitive
- average cost
- partially observable
- machine learning
- average reward
- planning under uncertainty
- basis functions
- action sets
- state abstraction
- learning algorithm
- sufficient conditions