Efficient Learning of a Linear Dynamical System with Stability Guarantees
Abstract
We propose a principled method for projecting an arbitrary square matrix to the non-convex set of asymptotically stable matrices. Leveraging ideas from large deviations theory, we show that this projection is optimal in an information-theoretic sense and that it simply amounts to shifting the initial matrix by an optimal linear quadratic feedback gain, which can be computed exactly and highly efficiently by solving a standard linear quadratic regulator problem. The proposed approach allows us to learn the system matrix of a stable linear dynamical system from a single trajectory of correlated state observations. The resulting estimator is guaranteed to be stable and offers explicit statistical bounds on the estimation error.
Cite
@article{arxiv.2102.03664,
title = {Efficient Learning of a Linear Dynamical System with Stability Guarantees},
author = {Wouter Jongeneel and Tobias Sutter and Daniel Kuhn},
journal= {arXiv preprint arXiv:2102.03664},
year = {2023}
}
Comments
Exposition has been updated