Online learning as an LQG optimal control problem with random matrices.

Giorgio Gnecco Alberto Bemporad Marco Gori Rita Morisi Marcello Sanguineti

Published in: ECC (2015)

Keyphrases

optimal control
online learning
feedback control
control problems
dynamic programming
infinite horizon
control strategy
risk sensitive
class of nonlinear systems
reinforcement learning
linear quadratic
control law
e learning
active learning
optimal control problems
multiscale