α-min: A Compact Approximate Solver For Finite-Horizon POMDPs.
Yann DujardinTom DietterichIadine ChadesPublished in: IJCAI (2015)
Keyphrases
- finite horizon
- markov decision processes
- optimal policy
- infinite horizon
- expected reward
- partially observable markov decision processes
- optimal stopping
- markov decision process
- partially observable
- reinforcement learning
- inventory models
- inventory control
- dynamic programming
- finite state
- single product
- point based value iteration
- state space
- multistage
- decision problems
- average cost
- long run
- yield management
- single item
- lost sales
- optimal control
- markov decision problems
- state dependent
- belief state
- action space
- production planning