Sparse estimation in linear dynamic networks using the stable spline horseshoe prior

DOI10.1016/j.automatica.2022.110666zbMath1505.93048OpenAlexW4307154503MaRDI QIDQ2097847

Gianluigi Pillonetto, Akram Yazdani

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.2022.110666

zbMATH Keywords

kernel-based regularizationlinear system identificationhorseshoe priorgroup-sparse estimationlinear dynamic networksstable spline kernel

Mathematics Subject Classification ID

System identification (93B30) Linear systems in control theory (93C05) Large-scale systems (93A15) Networked control (93B70)

Uses Software

• ElemStatLearn
• Hybrid Stable Spline Toolbox

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Maximum entropy properties of discrete-time first-order stable spline kernel
• Sparse network identifiability via compressed sensing
• Regularized linear system identification using atomic, nuclear and kernel-based norms: the role of the stability constraint
• A new kernel-based approach to hybrid system identification
• Kernel methods in system identification, machine learning and function estimation: a survey
• The horseshoe estimator: posterior concentration around nearly black vectors
• Prediction error identification of linear systems: a nonparametric Gaussian regression approach
• The quest for the right kernel in Bayesian impulse response identification: the use of OBFs
• A new kernel-based approach for linear system identification
• Open-loop asymptotically efficient model reduction with the Steiglitz-McBride method
• Identifiability of linear dynamic networks
• The power grid as a complex network: a survey
• Prediction error identification of linear dynamic networks with rank-reduced noise
• Least angle regression. (With discussion)
• High precision variational Bayesian inference of sparse linear networks
• On the mathematical foundations of stable RKHSs
• Identification of dynamic models in complex networks with prediction error methods-basic methods for consistent module estimates
• On the estimation of transfer functions, regularizations and Gaussian processes-revisited
• A Bayesian approach to sparse dynamic network identification
• On the half-Cauchy prior for a global scale parameter
• Dynamic network reconstruction from heterogeneous datasets
• Learning linear modules in a dynamic network using regularized kernel-based methods
• Identification of Dynamic Models in Complex Networks With Prediction Error Methods: Predictor Input Selection
• System Identification Via Sparse Multiple Kernel-Based Regularization Using Sequential Convex Optimization Techniques
• Kernel Absolute Summability Is Sufficient but Not Necessary for RKHS Stability
• The horseshoe estimator for sparse signals
• The Bayesian Lasso
• 10.1162/15324430152748236
• Necessary and Sufficient Conditions for Dynamical Structure Reconstruction of LTI Networks
• Topological Identification in Networks of Dynamical Systems
• A Local Direct Method for Module Identification in Dynamic Networks With Correlated Noise
• Identifiability of Dynamical Networks With Partial Node Measurements
• Theory of Reproducing Kernels
• An invariant form for the prior probability in estimation problems
• Signal Selection for Estimation and Identification in Networks of Dynamic Systems: A Graphical Model Approach
• The elements of statistical learning. Data mining, inference, and prediction