Uniqueness and Stability of Optimal Policies of Finite State Markov Decision Processes.
Arie LeizarowitzAlexander J. ZaslavskiPublished in: Math. Oper. Res. (2007)
Keyphrases
- finite state
- markov decision processes
- optimal policy
- sufficient conditions
- state space
- decision problems
- finite horizon
- dynamic programming
- policy iteration
- reinforcement learning
- infinite horizon
- average cost
- long run
- action sets
- average reward
- partially observable
- markov decision process
- control policies
- multistage
- state dependent
- initial state
- partially observable markov decision processes
- lost sales
- state and action spaces
- reinforcement learning algorithms
- model checking
- action space
- least squares
- supply chain
- total reward
- semi markov decision processes