Sample-path and variance minimization of Markov control processes with average cost criteria.
Onésimo Hernández-LermaOscar Vega-AmayaGuadalupe CarrascoPublished in: CDC (2000)
Keyphrases
- sample path
- average cost
- markov chain
- policy iteration
- finite state
- optimal control
- markov decision processes
- long run
- asymptotic analysis
- average reward
- inventory models
- infinite horizon
- optimal policy
- finite number
- finite horizon
- steady state
- linear programming
- control strategy
- model free
- lost sales
- multistage
- reinforcement learning
- transition probabilities
- state space
- stationary distribution
- control system
- fixed point
- cost function
- stationary points