Satisfiability Bounds for ω-Regular Properties in Bounded-Parameter Markov Decision Processes.
Jan KretínskýTobias MeggendorferMaximilian WeiningerPublished in: CoRR (2022)
Keyphrases
- markov decision processes
- state space
- optimal policy
- finite state
- reinforcement learning
- policy iteration
- discounted reward
- transition matrices
- dynamic programming
- infinite horizon
- decision theoretic planning
- finite horizon
- risk sensitive
- reachability analysis
- factored mdps
- average reward
- partially observable
- upper bound
- planning under uncertainty
- lower bound
- decision processes
- stochastic shortest path
- model based reinforcement learning
- action sets
- action space
- average cost
- reinforcement learning algorithms
- np complete
- computational complexity
- reward function
- decision problems
- machine learning