Discrete-Time Nonlinear Generalized Policy Iteration for Optimal Control Using Neural Networks.
Qinglai WeiDerong LiuXiong YangPublished in: ICONIP (1) (2014)
Keyphrases
- optimal control
- policy iteration
- optimal control problems
- linear quadratic
- neural network
- infinite horizon
- reinforcement learning
- dynamic programming
- control problems
- feedback control
- continuous stirred tank reactor
- approximate dynamic programming
- control strategy
- markov decision processes
- control law
- policy evaluation
- fixed point
- solving nonlinear
- actor critic
- average reward
- average cost
- optimal policy
- markov decision process
- model free
- artificial neural networks
- graphical models
- finite state
- function approximation
- differential evolution