Computing a Classic Index for Finite-Horizon Bandits.
José Niño-MoraPublished in: INFORMS J. Comput. (2011)
Keyphrases
- finite horizon
- optimal policy
- infinite horizon
- optimal stopping
- markov decision processes
- inventory models
- multistage
- inventory control
- markov decision process
- single product
- dynamic programming
- single item
- yield management
- lot size
- average cost
- machine learning
- sufficient conditions
- markov chain
- reinforcement learning