Complexity of Finite-Horizon Markov Decision process Problems.
Martin MundhenkJudy GoldsmithChristopher LusenaEric AllenderPublished in: Universität Trier, Mathematik/Informatik, Forschungsbericht (1999)
Keyphrases
- markov decision process
- finite horizon
- optimal policy
- infinite horizon
- markov decision processes
- state space
- optimal stopping
- reinforcement learning
- decision problems
- inventory models
- initial state
- inventory control
- multistage
- dynamic programming
- non stationary
- single product
- computational complexity
- finite state
- reward function
- average cost
- linear programming
- action space
- state dependent