Continuous-Time Markov Decision Processes Under the Risk-Sensitive First Passage Discounted Cost Criterion.
Qingda WeiXian ChenPublished in: J. Optim. Theory Appl. (2023)
Keyphrases
- risk sensitive
- markov decision processes
- average cost
- state space
- optimal control
- finite state
- optimal policy
- infinite horizon
- stationary policies
- finite horizon
- reinforcement learning
- dynamic programming
- markov decision process
- average reward
- partially observable
- policy iteration
- reinforcement learning algorithms
- action space
- optimality criterion
- total cost
- initial state
- long run
- control policy
- planning under uncertainty
- discounted reward
- expected cost
- decision processes
- decision making
- finite number
- control policies
- markov chain
- decision makers