Verification of general Markov decision processes by approximate similarity relations and policy refinement.
Sofie HaesaertSadegh Esmaeil Zadeh SoudjaniAlessandro AbatePublished in: CoRR (2016)
Keyphrases
- markov decision processes
- optimal policy
- policy iteration
- policy evaluation
- markov decision process
- state space
- infinite horizon
- factored mdps
- average reward
- finite horizon
- similarity relations
- reward function
- reinforcement learning
- average cost
- decision processes
- state and action spaces
- finite state
- dynamic programming
- action space
- transition matrices
- partially observable
- long run
- decision theoretic planning
- partially observable markov decision processes
- action sets
- expected reward
- continuous state spaces
- discounted reward
- sufficient conditions
- markov decision problems
- decision makers
- least squares
- integrity constraints
- artificial intelligence
- decision problems
- model free
- data mining