Trace Refinement in Labelled Markov Decision Processes.
Nathanaël FijalkowStefan KieferMahsa ShirmohammadiPublished in: FoSSaCS (2016)
Keyphrases
- markov decision processes
- reinforcement learning
- optimal policy
- state space
- finite state
- dynamic programming
- reinforcement learning algorithms
- transition matrices
- policy iteration
- average cost
- reachability analysis
- finite horizon
- model based reinforcement learning
- markov decision process
- average reward
- partially observable
- decision theoretic planning
- planning under uncertainty
- decision processes
- decision making
- state and action spaces
- factored mdps
- risk sensitive
- infinite horizon
- action space
- dynamical systems
- reward function
- multistage
- partially observable markov decision processes