Sample Complexity Bounds for Linear System Identification from a Finite Set
Not on the map. Why?
Abstract (arXiv)
This paper considers a finite sample perspective on the problem of identifying an LTI system from a finite set of possible systems using trajectory data. To this end, we use the maximum likelihood estimator to identify the true system and provide an upper bound for its sample complexity. Crucially, the derived bound does not rely on a potentially restrictive stability assumption. Additionally, we leverage tools from information theory to provide a lower bound to the sample complexity that holds independently of the used estimator. The derived sample complexity bounds are analyzed analytically and numerically.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 10 of its 23 references that are in the map. First posted on arXiv 17 Sep 2024.
Builds on
10 of 23 references.
References · 10
- 2017On the Sample Complexity of the Linear Quadratic Regulatorcited by 159
- 2018Learning Without Mixing: Towards A Sharp Analysis of Linear System Identificationcited by 158
- 2015Multi-armed Bandit Models for the Optimal Design of Clinical Trials: Benefits and Challengescited by 114
- 2018Near optimal finite time identification of arbitrary linear dynamical systemscited by 89
- 2016Explore First, Exploit Next: The True Shape of Regret in Bandit Problemscited by 79
- 2022Statistical Learning Theory for Control: A Finite Sample Perspectivecited by 53
- 2019Sample Complexity Lower Bounds for Linear System Identificationcited by 20
- 2021Linear Systems can be Hard to Learncited by 20
- 2019A Tutorial on Concentration Bounds for System Identificationcited by 19
- 2020Minimax Adaptive Control for a Finite Set of Linear Systemscited by 4
Built on by
Citing papers per year: 2025: 3.
3 papers.