Finite horizon continuous-time Markov decision processes with mean and variance criteria.
Yonghui HuangPublished in: Discret. Event Dyn. Syst. (2018)
Keyphrases
- finite horizon
- markov decision processes
- state space
- optimal policy
- infinite horizon
- optimal stopping
- reinforcement learning
- finite state
- policy iteration
- markov decision process
- average cost
- stationary policies
- markov chain
- dynamic programming
- partially observable
- transition matrices
- optimal control
- decision theoretic planning
- factored mdps
- standard deviation
- dynamical systems
- decision problems
- action space
- random walk
- decision making
- np hard
- control policies
- multistage
- optimal solution
- expected reward