Online Reinforcement Learning Control of Nonlinear Dynamic Systems: A State-action Value Function Based Solution.
Hamed Jabbari AslEiji UchibePublished in: Neurocomputing (2023)
Keyphrases
- state action
- reinforcement learning
- function approximators
- evaluation function
- nonlinear dynamic systems
- action space
- policy gradient
- average reward
- optimal control
- markov decision process
- function approximation
- state space
- neural network
- stochastic games
- action selection
- machine learning
- temporal difference
- markov decision processes
- learning algorithm
- nonlinear systems
- reward function
- kernel matrix
- state transitions
- recurrent neural networks
- transfer learning
- adaptive control
- control method
- control strategy
- evolutionary algorithm