Progressive State Space Disaggregation for Infinite Horizon Dynamic Programming.
Orso ForghieriHind CastelEmmanuel HyonErwan Le PennecPublished in: ICAPS (2024)
Keyphrases
- infinite horizon
- state space
- dynamic programming
- optimal policy
- markov decision processes
- finite horizon
- optimal control
- markov decision process
- heuristic search
- single item
- partially observable
- production planning
- stochastic demand
- average cost
- state variables
- markov chain
- planning problems
- multistage
- action space
- markov decision problems
- reinforcement learning
- linear programming
- reinforcement learning algorithms
- long run
- dynamical systems
- holding cost
- dec pomdps
- policy iteration
- decision problems
- state dependent
- lost sales
- real time
- inventory level
- fixed cost
- inventory control