Risk-sensitive discounted cost criterion for continuous-time Markov decision processes on a general state space.
Subrata GoluiChandan PalPublished in: Math. Methods Oper. Res. (2022)
Keyphrases
- markov decision processes
- risk sensitive
- state space
- average cost
- optimal policy
- reinforcement learning
- dynamic programming
- finite state
- finite horizon
- markov chain
- planning under uncertainty
- partially observable
- heuristic search
- policy iteration
- state variables
- markov decision process
- average reward
- infinite horizon
- action space
- optimal control
- stationary policies
- markov decision problems
- decision processes
- reinforcement learning algorithms
- initial state
- state and action spaces
- discounted reward
- optimality criterion
- belief state
- reward function
- planning problems
- dynamical systems
- multistage
- total cost
- least squares
- search space