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.
System identification
5 Pith papers cite this work. Polarity classification is still indexing.
5
Pith papers citing it
years
2026 5verdicts
UNVERDICTED 5representative citing papers
Informativity conditions enable unique output prediction for LTI systems from data without uniquely identifying the underlying dynamics.
Drifting MPC produces a unique distribution over trajectories that trades off data support against optimality and enables efficient receding-horizon planning under unknown dynamics.
Develops a two-stage system identification plus sensor allocation algorithm with non-asymptotic guarantees for near-optimal sensor counts in unknown high-dimensional linear systems.
A mixed-integer quadratically constrained optimization learns interpretable stable dynamical models and their Lyapunov functions from data by enforcing stability constraints during training.
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.
-
Informativity for Data-driven Prediction
Informativity conditions enable unique output prediction for LTI systems from data without uniquely identifying the underlying dynamics.
-
Receding-Horizon Control via Drifting Models
Drifting MPC produces a unique distribution over trajectories that trades off data support against optimality and enables efficient receding-horizon planning under unknown dynamics.
-
Provably Efficient Sensor Allocation for Unknown High-dimensional Systems with Limited Sensing
Develops a two-stage system identification plus sensor allocation algorithm with non-asymptotic guarantees for near-optimal sensor counts in unknown high-dimensional linear systems.
-
Learning interpretable and stable dynamical models via mixed-integer Lyapunov-constrained optimization
A mixed-integer quadratically constrained optimization learns interpretable stable dynamical models and their Lyapunov functions from data by enforcing stability constraints during training.