Improved rates for prediction and identification for partially observed linear dynamical systems
Lists
Only you see this
Abstract
Identification of a linear time-invariant dynamical system from partial observations is a fundamental problem in control theory. Particularly challenging are systems exhibiting long-term memory. A natural question is how learn such systems with non-asymptotic statistical rates depending on the inherent dimensionality (order) $d$ of the system, rather than on the possibly much larger memory length. We propose an algorithm that given a single trajectory of length $T$ with gaussian observation noise, learns the system with a near-optimal rate of $\widetilde O\left(\sqrt\frac{d}{T}\right)$ in $\mathcal{H}_2$ error, with only logarithmic, rather than polynomial dependence on memory length. We also give bounds under process noise and improved bounds for learning a realization of the system. Our algorithm is based on multi-scale low-rank approximation: SVD applied to Hankel matrices of geometrically increasing sizes. Our analysis relies on careful application of concentration bounds on the Fourier domain -- we give sharper concentration bounds for sample covariance of correlated inputs and for $\cal H_\infty$ norm estimation, which may be of independent interest.
BibTeX
@inproceedings{lee2021improved,
author = {Holden Lee},
title = {{Improved rates for prediction and identification for partially observed linear dynamical systems}},
booktitle = {Advances in Neural Information Processing Systems},
year = {2021},
url = {https://openreview.net/forum?id=uJDmJp7kn2F},
}