Risk Probability Minimization Problems for Continuous-Time Markov Decision Processes on Finite Horizon.
Haifeng HuoXianping GuoPublished in: IEEE Trans. Autom. Control. (2020)
Keyphrases
- markov decision processes
- finite horizon
- minimization problems
- state space
- optimal policy
- expected reward
- stationary policies
- infinite horizon
- total variation
- average cost
- finite state
- reinforcement learning
- markov decision process
- markov chain
- partially observable
- action space
- dynamic programming
- low rank
- policy iteration
- probability distribution
- average reward
- lost sales
- optimal control
- posterior probability
- dynamical systems
- transition probabilities
- multistage
- pairwise
- reward function
- initial state
- learning algorithm
- least squares
- search space
- decision making