Derives an ε-local minimax excess-cost lower bound for learning LQG controllers from offline trajectories of a linear exploration policy, expressed via the Hessian of the LQG cost and inverse Fisher information, and instantiates it on fragile robust-control examples.
High effort, low gain: Fundamental limits of active learning for linear dynamical systems
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
An active learning algorithm for linear systems attains the minimal sample complexity for accurate identification using ordinary least squares and semidefinite programming with centered excitation.
citing papers explorer
-
The Fragility of Learning LQG Controllers
Derives an ε-local minimax excess-cost lower bound for learning LQG controllers from offline trajectories of a linear exploration policy, expressed via the Hessian of the LQG cost and inverse Fisher information, and instantiates it on fragile robust-control examples.
-
Optimal Centered Active Excitation in Linear System Identification
An active learning algorithm for linear systems attains the minimal sample complexity for accurate identification using ordinary least squares and semidefinite programming with centered excitation.