Regret Bounds for Model-Free Linear Quadratic Control.

Yasin Abbasi-Yadkori Nevena Lazic Csaba Szepesvári

Published in: CoRR (2018)

Keyphrases

model free
linear quadratic
optimal control
reinforcement learning
closed loop
vector valued
function approximation
control system
dynamical systems
machine learning
control strategy
gaussian model
policy iteration
neural network
feature vectors
nearest neighbor