Lightweight Verification of Markov Decision Processes with Rewards.
Axel LegaySean SedwardsLouis-Marie TraonouezPublished in: CoRR (2014)
Keyphrases
- markov decision processes
- lightweight
- reinforcement learning
- optimal policy
- state space
- finite state
- policy iteration
- transition matrices
- dynamic programming
- finite horizon
- reward function
- sequential decision making under uncertainty
- model checking
- infinite horizon
- decision theoretic planning
- factored mdps
- planning under uncertainty
- reinforcement learning algorithms
- action space
- average cost
- markov decision process
- partially observable
- reachability analysis
- risk sensitive
- decision processes
- state and action spaces
- development environments
- communication infrastructure
- total reward
- model based reinforcement learning
- machine learning
- markov decision problems
- action sets
- decision makers
- interval estimation