Continuous-time Markov decision processes with risk-sensitive finite-horizon cost criterion.
Qingda WeiPublished in: Math. Methods Oper. Res. (2016)
Keyphrases
- risk sensitive
- finite horizon
- markov decision processes
- average cost
- state space
- optimal control
- optimal policy
- infinite horizon
- finite state
- dynamic programming
- reinforcement learning
- control policies
- markov chain
- average reward
- long run
- lost sales
- markov decision process
- dynamical systems
- optimality criterion
- initial state
- reinforcement learning algorithms
- partially observable
- decision processes
- policy iteration
- finite number
- markov decision problems
- linear programming
- multistage
- expected reward
- expected cost
- reward function
- total cost
- linear program
- non stationary