Eventually-stationary policies for Markov decision models with non-constant discounting.
Yair CarmonAdam ShwartzPublished in: VALUETOOLS (2008)
Keyphrases
- decision models
- stationary policies
- markov decision processes
- decision model
- markov chain
- finite state
- state space
- influence diagrams
- markov decision process
- optimal policy
- decision theoretic
- lot sizing
- decision problems
- average cost
- dynamic programming
- sufficient conditions
- linear program
- model construction
- learning algorithm
- markov decision problems
- expected utility
- long run
- markov model
- multistage
- probabilistic reasoning
- initial state
- infinite horizon
- evolutionary algorithm
- finite number
- sensitivity analysis
- linear programming
- decision making
- search algorithm
- lower bound
- supervised learning