Statistically Model Checking PCTL Specifications on Markov Decision Processes via Reinforcement Learning.
Yu WangNima RoohiMatthew WestMahesh ViswanathanGeir E. DullerudPublished in: CDC (2020)
Keyphrases
- markov decision processes
- model checking
- reinforcement learning
- automated verification
- model checker
- formal specification
- transition systems
- finite state
- bounded model checking
- reachability analysis
- formal verification
- temporal logic
- optimal policy
- reactive systems
- reinforcement learning algorithms
- concurrent systems
- finite state machines
- state space
- state and action spaces
- policy iteration
- action space
- model based reinforcement learning
- symbolic model checking
- formal methods
- markov decision process
- decision theoretic planning
- dynamic programming
- timed automata
- average reward
- infinite horizon
- partially observable
- machine learning
- learning algorithm
- asynchronous circuits
- function approximation
- state abstraction
- decentralized control
- planning under uncertainty
- average cost
- reward function
- model free
- description language
- optimal control
- fully observable
- belief revision
- decision problems