Dynamic Programming-based Approximate Optimal Control for Model-Based Reinforcement Learning.
Prakash MallickZhiyong ChenPublished in: CoRR (2023)
Keyphrases
- optimal control
- dynamic programming
- model based reinforcement learning
- markov decision processes
- reinforcement learning
- control problems
- state space
- infinite horizon
- multistage
- optimal policy
- optimal control problems
- finite state
- knapsack problem
- partially observable markov decision processes
- policy iteration
- average cost
- linear programming
- decision processes
- control strategy
- learning algorithm
- reward function
- partially observable
- machine learning