Sample complexity of the sign-perturbed sums method

DOI10.1016/J.AUTOMATICA.2024.112020MaRDI QIDQ6964502zbMATH OpenFDO

Authors Szabolcs Szentpéteri, Balázs Csanád Csáji

Publication date 4 July 2025

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/J.AUTOMATICA.2024.112020

zbMATH Keywords

confidence regionsfinite sample propertiesleast squares methodlinear regressionsample complexity

Mathematics Subject Classification ID

Linear systems in control theory (93C05) Identification in stochastic control theory (93E12) Least squares and related methods for stochastic control systems (93E24)

Cites work

• A simple condition for the boundedness of sign-perturbed-sums (SPS) confidence regions
• Asymptotic properties of SPS confidence regions
• Bayesian frequentist bounds for machine learning and system identification
• Exact matrix completion via convex optimization
• Finite sample performance of linear least squares estimation
• Finite time LTI system identification
• Guaranteed characterization of exact non-asymptotic confidence regions as defined by LSCR and SPS
• Guaranteed non-asymptotic confidence regions in system identification
• High-dimensional statistics. A non-asymptotic viewpoint
• Non-asymptotic state-space identification of closed-loop stochastic linear systems using instrumental variables
• Perturbed datasets methods for hypothesis testing and structure of corresponding confidence sets
• Revisiting Ho–Kalman-Based System Identification: Robustness and Finite-Sample Analysis
• Sign-Perturbed Sums: A New System Identification Approach for Constructing Exact Non-Asymptotic Confidence Regions in Linear Regression Models
• The sample complexity of learning linear predictors with the squared loss

This page was built for publication: Sample complexity of the sign-perturbed sums method

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6964502)