Neural approximations in discounted infinite-horizon stochastic optimal control problems.
Giorgio GneccoMarcello SanguinetiPublished in: Eng. Appl. Artif. Intell. (2018)
Keyphrases
- infinite horizon
- optimal control problems
- optimal control
- production planning
- finite horizon
- dynamic programming
- single item
- stochastic demand
- average cost
- control strategy
- network architecture
- partially observable
- optimal policy
- markov decision processes
- long run
- reinforcement learning
- neural network
- fixed cost
- markov decision process
- state space
- inventory control
- control law
- periodic review