Sparse Volterra and Polynomial Regression Models: Recoverability and Estimation

DOI10.1109/TSP.2011.2165952zbMATH Open1393.94687arXiv1103.0769MaRDI QIDQ4573365FDOQ4573365

Authors: Vassilis Kekatos, Georgios B. Giannakis 

Publication date: 18 July 2018

Published in: IEEE Transactions on Signal Processing (Search for Journal in Brave)

Abstract: Volterra and polynomial regression models play a major role in nonlinear system identification and inference tasks. Exciting applications ranging from neuroscience to genome-wide association analysis build on these models with the additional requirement of parsimony. This requirement has high interpretative value, but unfortunately cannot be met by least-squares based or kernel regression methods. To this end, compressed sampling (CS) approaches, already successful in linear regression settings, can offer a viable alternative. The viability of CS for sparse Volterra and polynomial models is the core theme of this work. A common sparse regression task is initially posed for the two models. Building on (weighted) Lasso-based schemes, an adaptive RLS-type algorithm is developed for sparse polynomial regressions. The identifiability of polynomial models is critically challenged by dimensionality. However, following the CS principle, when these models are sparse, they could be recovered by far fewer measurements. To quantify the sufficient number of measurements for a given level of sparsity, restricted isometry properties (RIP) are investigated in commonly met polynomial regression settings, generalizing known results for their linear counterparts. The merits of the novel (weighted) adaptive CS algorithms to sparse polynomial modeling are verified through synthetic as well as real data tests for genotype-phenotype analysis.

Full work available at URL: https://arxiv.org/abs/1103.0769

Mathematics Subject Classification ID

Identification in stochastic control theory (93E12) Sampling theory in information and communication theory (94A20)

Cited In (6)

• Optimal modeling of nonlinear systems: method of variable injections
• Identification of Wiener-Hammerstein systems by \(\ell_1\)-constrained Volterra series
• Statistical inference in compound functional models
• Learning general sparse additive models from point queries in high dimensions
• Sparse hierarchical regression with polynomials
• Sparse estimation technique for digital pre-distortion of impedance-mismatched power amplifiers

This page was built for publication: Sparse Volterra and Polynomial Regression Models: Recoverability and Estimation

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4573365)