Characterization of the optimal average cost in Markov decision chains driven by a risk-seeking controller.
Rolando Cavazos-CadenaHugo Cruz-SuárezRaúl Montes-de-OcaPublished in: J. Appl. Probab. (2024)
Keyphrases
- markov decision chains
- average cost
- optimal control
- risk sensitive
- control policy
- long run
- markov decision processes
- finite number
- finite state
- finite horizon
- dynamic programming
- initial state
- infinite horizon
- optimality criterion
- control strategy
- optimal policy
- approximate dynamic programming
- multistage
- inventory models
- linear programming
- holding cost
- real time
- total cost
- linear program
- action sets
- decision making
- utility function
- control policies
- reinforcement learning
- setup cost
- markov decision process
- search space
- control system
- machine learning