Duality between large deviation control and risk-sensitive control for Markov decision processes.
Yanan DaiJinwen ChenPublished in: Syst. Control. Lett. (2023)
Keyphrases
- markov decision processes
- risk sensitive
- optimal control
- reinforcement learning
- control problems
- finite state
- state space
- control system
- large deviations
- control policies
- least squares
- reinforcement learning algorithms
- dynamic programming
- optimal policy
- markov random field
- action selection
- partially observable
- average cost