Zero-sum games for continuous-time Markov jump processes with risk-sensitive finite-horizon cost criterion.
Qingda WeiPublished in: Oper. Res. Lett. (2018)
Keyphrases
- risk sensitive
- finite horizon
- average cost
- markov chain
- finite state
- optimal control
- markov decision processes
- infinite horizon
- state space
- optimal policy
- control policies
- long run
- markov model
- steady state
- total cost
- fixed cost
- finite number
- markov decision problems
- transition probabilities
- markov decision process
- random walk
- optimality criterion
- dynamic programming
- control policy
- linear program
- multistage
- stochastic processes
- initial state
- decision processes
- inventory policy
- lot size
- policy iteration
- linear programming
- reinforcement learning
- average reward
- control strategy
- real time