Actor-Critic based Improper Reinforcement Learning.
Mohammadi ZakiAvi MohanAditya GopalanShie MannorPublished in: ICML (2022)
Keyphrases
- actor critic
- reinforcement learning
- temporal difference
- optimal control
- reinforcement learning algorithms
- policy gradient
- approximate dynamic programming
- neuro fuzzy
- gradient method
- function approximation
- policy iteration
- state space
- model free
- average reward
- control problems
- markov decision processes
- multi agent
- natural actor critic
- rl algorithms
- policy gradient methods
- learning algorithm
- action selection
- dynamical systems
- machine learning
- step size
- evaluation function
- linear program