Iterative Q-Learning for Model-Free Optimal Control With Adjustable Convergence Rate.
Ding WangYuan WangMingming ZhaoJunfei QiaoPublished in: IEEE Trans. Circuits Syst. II Express Briefs (2024)
Keyphrases
- model free
- convergence rate
- optimal control
- reinforcement learning
- reinforcement learning algorithms
- risk sensitive
- policy iteration
- function approximation
- learning rate
- control problems
- step size
- convergence speed
- dynamic programming
- global convergence
- temporal difference
- rl algorithms
- infinite horizon
- control law
- optimal control problems
- optimal policy
- state space
- impedance control
- control strategy
- gradient method
- supervised learning
- learning algorithm
- average reward
- policy evaluation
- average cost
- learning problems
- function approximators
- markov decision processes
- markov decision problems
- machine learning