Structured Policy Iteration for Linear Quadratic Regulator.
Youngsuk ParkRyan A. RossiZheng WenGang WuHandong ZhaoPublished in: ICML (2020)
Keyphrases
- linear quadratic
- policy iteration
- optimal control
- infinite horizon
- markov decision processes
- dynamical systems
- closed loop
- reinforcement learning
- vector valued
- dynamic programming
- fixed point
- optimal policy
- control strategy
- model free
- least squares
- state space
- finite state
- gaussian model
- markov decision process
- learning algorithm
- maximum likelihood
- temporal difference
- average cost
- linear programming
- control system