Compositional Synthesis of Interconnected Stochastic Control Systems based on Finite MDPs.
Abolfazl LavaeiSadegh SoudjaniMajid ZamaniPublished in: HSCC (2018)
Keyphrases
- control system
- markov decision processes
- state and action spaces
- control policies
- discrete random variables
- action space
- optimal policy
- reinforcement learning
- stochastic domains
- monte carlo
- factored mdps
- finite number
- finite horizon
- decision theoretic planning
- stochastic programming
- state space
- program synthesis
- continuous state spaces
- real environment
- markov decision process
- stochastic processes
- average cost
- decision diagrams
- closed loop
- decision problems
- dynamic programming
- real time dynamic programming
- search space