Notes on average Markov decision processes with a minimum-variance criterion.
Jianyong LiuPublished in: Oper. Res. Lett. (2002)
Keyphrases
- markov decision processes
- minimum variance
- average cost
- finite state
- state space
- optimal policy
- discounted reward
- transition matrices
- reinforcement learning
- dynamic programming
- policy iteration
- decision theoretic planning
- action space
- infinite horizon
- linear prediction
- stationary policies
- average reward
- portfolio optimization
- estimation error
- optimality criterion
- action sets