On Supervised Online Rolling-Horizon Control for Infinite-Horizon Discounted Markov Decision Processes.
Hyeong Soo ChangPublished in: IEEE Trans. Autom. Control. (2024)
Keyphrases
- infinite horizon
- markov decision processes
- rolling horizon
- optimal control
- finite horizon
- optimal policy
- average cost
- dynamic programming
- finite state
- state space
- single item
- policy iteration
- partially observable
- lot sizing
- markov decision process
- reinforcement learning
- average reward
- long run
- learning algorithm
- control strategy
- planning under uncertainty
- control policies
- decision processes
- stationary policies
- dec pomdps
- reinforcement learning algorithms
- action space
- reward function
- single machine
- state dependent
- optimal solution
- multi agent
- mathematical model
- scheduling problem
- control system