Successive Linear Approximation Solution of Infinite-Horizon Dynamic Stochastic Programs.
John R. BirgeGongyun ZhaoPublished in: SIAM J. Optim. (2007)
Keyphrases
- linear approximation
- infinite horizon
- policy iteration
- finite horizon
- markov decision processes
- optimal policy
- long run
- optimal control
- dynamic programming
- markov decision process
- dynamic environments
- basis functions
- inventory policy
- optimal solution
- reinforcement learning
- partially observable
- least squares
- average reward
- inventory control
- bayesian networks
- stochastic demand
- machine learning