Adaptive Gaussian inverse regression with partially unknown operator

DOI10.1080/03610926.2012.731548MaRDI QIDQ4929192zbMATH OpenOpenAlexFDO

Authors Jan Johannes, Maik Schwarz

Publication date 13 June 2013

Published in Communications in Statistics: Theory and Methods (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1204.1226

zbMATH Keywords

model selectionSobolev spacesLepski's methodminimax theoryadaptive nonparametric estimationGaussian sequence space modelmildly and severely ill-posed inverse problems

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Nonparametric regression and quantile regression (62G08) Minimax procedures in statistical decision theory (62C20)

Abstract: This work deals with the ill-posed inverse problem of reconstructing a function $f$ given implicitly as the solution of $g=Af$, where $A$ is a compact linear operator with unknown singular values and known eigenfunctions. We observe the function $g$ and the singular values of the operator subject to Gaussian white noise with respective noise levels $varepsilon$ and $sigma$. We develop a minimax theory in terms of both noise levels and propose an orthogonal series estimator attaining the minimax rates. This estimator requires the optimal choice of a dimension parameter depending on certain characteristics of $f$ and $A$. This work addresses the fully data-driven choice of the dimension parameter combining model selection with Lepski's method. We show that the fully data-driven estimator preserves minimax optimality over a wide range of classes for $f$ and $A$ and noise levels $varepsilon$ and $sigma$. The results are illustrated considering Sobolev spaces and mildly and severely ill-posed inverse problems.

Recommendations

• Adaptive estimation for inverse problems with noisy operators
• Empirical risk minimization in inverse problems
• Oracle inequalities for inverse problems
• On inverse problems with unknown operators
• Rate optimal estimation of quadratic functionals in inverse problems with partially unknown operator and application to testing problems

Cites work

• scientific article; zbMATH DE number 741240 (Why is no real title available?)
• Adaptive estimation for inverse problems with noisy operators
• Bandwidth selection in kernel density estimation: oracle inequalities and adaptive minimax optimality
• Deconvolving kernel density estimators
• Nonlinear estimation for linear inverse problems with error in the operator
• Nonparametric quasi-maximum likelihood estimation for Gaussian locally stationary processes
• On inverse problems with unknown operators
• On optimal solutions of the deconvolution problem
• On the effect of estimating the error density in nonparametric deconvolution
• On the optimal rates of convergence for nonparametric deconvolution problems
• Optimal discretization of inverse problems in Hilbert scales. Regularization and self-regulari\-za\-tion of projection methods
• Oracle inequalities for inverse problems
• Risk bounds for model selection via penalization
• Speed of estimation in positron emission tomography and related inverse problems
• Statistical Inverse Estimation in Hilbert Scales

Cited in

(9)

• Deep learning for inverse problems with unknown operator
• On inverse problems with unknown operators
• Estimation of convolution in the model with noise
• An adaptive ANOVA-based PCKF for high-dimensional nonlinear inverse modeling
• Adaptive estimation for an inverse regression model with unknown operator
• Adaptive estimation for inverse problems with noisy operators
• Nonparametric intensity estimation from noisy observations of a Poisson process under unknown error distribution
• Rate optimal estimation of quadratic functionals in inverse problems with partially unknown operator and application to testing problems
• Minimax goodness-of-fit testing in ill-posed inverse problems with partially unknown operators

This page was built for publication: Adaptive Gaussian inverse regression with partially unknown operator

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4929192)