Reinforcement Learning with Intrinsically Motivated Feedback Graph for Lost-sales Inventory Control.
Zifan LiuXinran LiShibo ChenGen LiJiashuo JiangJun ZhangPublished in: CoRR (2024)
Keyphrases
- inventory control
- lost sales
- optimal policy
- reinforcement learning
- inventory models
- inventory systems
- stochastic demand
- finite horizon
- periodic review
- supply chain
- markov decision processes
- lead time
- state space
- decision problems
- base stock policies
- inventory level
- infinite horizon
- single stage
- long run
- customer service
- dynamic programming
- random walk
- multi item
- markov decision process
- multistage
- dynamic pricing
- setup cost
- state dependent
- average cost
- demand distributions
- single item
- initial state
- finite state
- special case