Strategy Synthesis in Markov Decision Processes Under Limited Sampling Access.
Christel BaierClemens DubslaffPatrick WienhöftStefan J. KiebelPublished in: NFM (2023)
Keyphrases
- markov decision processes
- finite state
- optimal policy
- state space
- transition matrices
- reachability analysis
- decision theoretic planning
- reinforcement learning
- policy iteration
- dynamic programming
- decision processes
- partially observable
- state and action spaces
- factored mdps
- risk sensitive
- average reward
- planning under uncertainty
- reinforcement learning algorithms
- model based reinforcement learning
- average cost
- action space
- sample size
- finite horizon
- action sets
- infinite horizon
- learning algorithm
- state abstraction
- optimal strategy