Bayesian frequentist bounds for machine learning and system identification

DOI10.1016/J.AUTOMATICA.2022.110599MaRDI QIDQ2097759zbMATH OpenOpenAlexFDO

Authors Giacomo Baggio, A. Carè, Anna Scampicchio, Gianluigi Pillonetto

Publication date 14 November 2022

Published in Automatica (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1016/j.automatica.2022.110599

zbMATH Keywords

finite sample system identificationGaussian regressionkernel-based non-parametric methodssystem identificationuncertainty quantification

Mathematics Subject Classification ID

Learning and adaptive systems in artificial intelligence (68T05) Identification in stochastic control theory (93E12)

Recommendations

• Bounds from noisy linear measurements
• Bayesian differential programming for robust systems identification under uncertainty
• scientific article; zbMATH DE number 4204258
• On Bayes risk lower bounds
• Deterministic error bounds for kernel-based learning techniques under bounded noise

Cites work

• A Correspondence Between Bayesian Estimation on Stochastic Processes and Smoothing by Splines
• A new kernel-based approach for linear system identification
• A reproducing kernel Hilbert space approach to functional linear regression
• An $\ell _{1}$-Laplace Robust Kalman Smoother
• Asymptotic properties of SPS confidence regions
• Bayes Factors
• Concentration estimates for learning with unbounded sampling
• Distribution-free uncertainty quantification for kernel methods by gradient perturbations
• Facing undermodelling in sign-perturbed-sums system identification
• Gaussian processes for machine learning.
• Generalized Cross-Validation as a Method for Choosing a Good Ridge Parameter
• Guaranteed non-asymptotic confidence regions in system identification
• scientific article; zbMATH DE number 5957408 (Why is no real title available?)
• scientific article; zbMATH DE number 44406 (Why is no real title available?)
• scientific article; zbMATH DE number 45848 (Why is no real title available?)
• scientific article; zbMATH DE number 1332320 (Why is no real title available?)
• scientific article; zbMATH DE number 1843268 (Why is no real title available?)
• scientific article; zbMATH DE number 1911984 (Why is no real title available?)
• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• scientific article; zbMATH DE number 849926 (Why is no real title available?)
• Kernel methods in system identification, machine learning and function estimation: a survey
• Learning rates of least-square regularized regression
• Learning theory estimates via integral operators and their approximations
• Matrix theory. Basic results and techniques
• Non-asymptotic confidence regions for regularized linear regression estimates
• Nonparametric estimation of transfer functions: rates of convergence and adaptation
• On the sample complexity of the linear quadratic regulator
• Optimal learning rates for least squares regularized regression with unbounded sampling
• Optimal rates for regularization of statistical inverse learning problems
• Particle Markov Chain Monte Carlo Methods
• Regularization networks and support vector machines
• Robust Statistics
• Scalable approximate MCMC algorithms for the horseshoe prior
• Sign-Perturbed Sums: A New System Identification Approach for Constructing Exact Non-Asymptotic Confidence Regions in Linear Regression Models
• The Bayesian Bridge
• The Bayesian Lasso
• The elements of statistical learning. Data mining, inference, and prediction
• The interplay of Bayesian and frequentist analysis
• Theory of Reproducing Kernels
• Universal kernels
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties

Cited in

(7)

• Bayesian positive system identification: truncated Gaussian prior and hyperparameter estimation
• Scalable bounding of predictive uncertainty in regression problems with SLAC
• Uncertainty quantification in neural network classifiers -- a local linear approach
• Deep networks for system identification: a survey
• Sample complexity of the sign-perturbed sums method
• Signed-perturbed sums estimation of ARX systems: exact coverage and strong consistency
• A highly-accurate identification method for linear systems using transferred knowledge

Describes a project that uses

Uses Software

• Gibbsit
• BayesDA
• ElemStatLearn

This page was built for publication: Bayesian frequentist bounds for machine learning and system identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2097759)