Risk hull method and regularization by projections of ill-posed inverse problems

DOI10.1214/009053606000000542zbMATH Open1246.62082arXivmath/0611228OpenAlexW2034772703MaRDI QIDQ449941FDOQ449941

Authors: Laurent Cavalier, Yu. Golubev 

Publication date: 3 September 2012

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: We study a standard method of regularization by projections of the linear inverse problem $Y=Af+epsilon$, where $epsilon$ is a white Gaussian noise, and $A$ is a known compact operator with singular values converging to zero with polynomial decay. The unknown function $f$ is recovered by a projection method using the singular value decomposition of $A$. The bandwidth choice of this projection regularization is governed by a data-driven procedure which is based on the principle of risk hull minimization. We provide nonasymptotic upper bounds for the mean square risk of this method and we show, in particular, that in numerical simulations this approach may substantially improve the classical method of unbiased risk estimation.

Full work available at URL: https://arxiv.org/abs/math/0611228

Recommendations

• Risk hull method for spectral regularization in linear statistical inverse problems
• scientific article; zbMATH DE number 3917571
• On the regularization of projection methods for solving ill-posed problems
• scientific article; zbMATH DE number 3903868
• Risk estimators for choosing regularization parameters in ill-posed problems -- properties and limitations
• Projection methods for ill-posed problems revisited
• scientific article; zbMATH DE number 4054935
• scientific article; zbMATH DE number 3992815
• scientific article; zbMATH DE number 833247
• Incremental projection approach of regularization for inverse problems

zbMATH Keywords

oracle inequalityquadratic risk

Mathematics Subject Classification ID

Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Inference from stochastic processes (62M99)

Cites Work

• Title not available (Why is that?)
• Some Comments on C P
• Estimation of the mean of a multivariate normal distribution
• Nonlinear solution of linear inverse problems by wavelet-vaguelette decomposition
• Title not available (Why is that?)
• Gaussian model selection
• On the optimal rates of convergence for nonparametric deconvolution problems
• Risk bounds for model selection via penalization
• A statistical perspective on ill-posed inverse problems (with discussion)
• Title not available (Why is that?)
• Title not available (Why is that?)
• Oracle inequalities for inverse problems
• Sharp adaptation for inverse problems with random noise
• Title not available (Why is that?)
• Title not available (Why is that?)
• Speed of estimation in positron emission tomography and related inverse problems
• Integral equations. Theory and numerical treatment
• Statistical Inverse Estimation in Hilbert Scales
• Title not available (Why is that?)
• Robust and efficient recovery of a signal passed through a filter and then contaminated by non-Gaussian noise
• Block Thresholding and Sharp Adaptive Estimation in Severely Ill-Posed Inverse Problems
• The principle of penalized empirical risk in severely ill-posed problems
• An oracle approach to adaptive estimation of linear functionals in a Gaussian model

Cited In (40)

• Optimal Convergence of the Discrepancy Principle for Polynomially and Exponentially Ill-Posed Operators under White Noise
• Upper functions for \(\mathbb{L}_{p}\)-norms of Gaussian random fields
• Bayes procedures for adaptive inference in inverse problems for the white noise model
• Uncertainty quantification for robust variable selection and multiple testing
• Solution of linear ill-posed problems using overcomplete dictionaries
• Upper functions for positive random functionals. I: General setting and Gaussian random functions
• Comparing parameter choice methods for regularization of ill-posed problems
• Spatially inhomogeneous linear inverse problems with possible singularities
• Optimal regularized hypothesis testing in statistical inverse problems
• A new approach to estimator selection
• Testing monotonicity of pricing kernels
• Empirical risk minimization as parameter choice rule for general linear regularization methods
• Early stopping for statistical inverse problems via truncated SVD estimation
• Uniform bounds for norms of sums of independent random functions
• Multiscale scanning in inverse problems
• On oracle inequalities related to data-driven hard thresholding
• Tikhonov-Phillips regularizations in linear models with blurred design
• Solution of linear ill-posed problems by model selection and aggregation
• Adaptive estimation over anisotropic functional classes via oracle approach
• Ill-Posed Problems: Operator Methodologies of Resolution and Regularization
• Adaptivity and Oracle Inequalities in Linear Statistical Inverse Problems: A (Numerical) Survey
• On universal oracle inequalities related to high-dimensional linear models
• On the stability of the risk hull method for projection estimators
• Bayesian model selection and the concentration of the posterior of hyperparameters
• Optimal Adaptation for Early Stopping in Statistical Inverse Problems
• Operator-theoretic and regularization approaches to ill-posed problems
• The principle of penalized empirical risk in severely ill-posed problems
• Regularization parameter selection in indirect regression by residual based bootstrap
• Theory of adaptive estimation
• The empirical process of residuals from an inverse regression
• Semi‐parametric Estimation in a Single‐index Model with Endogenous Variables
• Spectral cut-off regularizations for ill-posed linear models
• Beyond the Bakushinkii veto: regularising linear inverse problems without knowing the noise distribution
• Risk hull method for spectral regularization in linear statistical inverse problems
• Adaptive complexity regularization for linear inverse problems
• Adaptive spectral regularizations of high dimensional linear models
• The Stein hull
• Sparse model selection under heterogeneous noise: exact penalisation and data-driven thresholding
• Minimax goodness-of-fit testing in ill-posed inverse problems with partially unknown operators
• On empirical Bayes approach to inverse problems

This page was built for publication: Risk hull method and regularization by projections of ill-posed inverse problems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q449941)