Trace Refinement in Labelled Markov Decision Processes.
Nathanaël FijalkowStefan KieferMahsa ShirmohammadiPublished in: Log. Methods Comput. Sci. (2020)
Keyphrases
- markov decision processes
- optimal policy
- state space
- transition matrices
- reinforcement learning
- finite state
- dynamic programming
- average cost
- reinforcement learning algorithms
- policy iteration
- reachability analysis
- planning under uncertainty
- decision theoretic planning
- average reward
- model based reinforcement learning
- finite horizon
- action space
- action sets
- markov decision process
- decision processes
- machine learning
- factored mdps
- partially observable
- risk sensitive
- stationary policies
- infinite horizon
- decision problems