Exploiting Linear Models for Model-Free Nonlinear Control: A Provably Convergent Policy Gradient Approach.
Guannan QuChenkai YuSteven H. LowAdam WiermanPublished in: CDC (2021)
Keyphrases
- model free
- linear models
- reinforcement learning algorithms
- policy gradient
- reinforcement learning
- provably convergent
- function approximation
- average reward
- linear model
- variable selection
- temporal difference
- linear regression
- reinforcement learning methods
- shape from shading
- optimal control
- adaptive control
- control strategy
- state space
- policy iteration
- rl algorithms
- control system
- feature vectors
- stochastic games
- decision trees