Adaptive minimax optimality in statistical inverse problems via SOLIT—Sharp Optimal Lepskiĭ-Inspired Tuning

DOI10.1088/1361-6420/AD12E0arXiv2304.10356OpenAlexW4389367969MaRDI QIDQ6136775FDOQ6136775

Housen Li, Frank Werner

Publication date: 17 January 2024

Published in: Inverse Problems (Search for Journal in Brave)

Abstract: We consider statistical linear inverse problems in separable Hilbert spaces and filter-based reconstruction methods of the form $hat{f}_{a}lpha=q_{a}lphaleft(T^{*}Tight)T^{*}Y$, where $Y$ is the available data, $T$ the forward operator, $left(q_{a}lphaight{)}_{alphainmathcalA}$ an ordered filter, and $alpha>0$ a regularization parameter. Whenever such a method is used in practice, $alpha$ has to be chosen appropriately. Typically, the aim is to find or at least approximate the best possible $alpha$ in the sense that mean squared error (MSE) $mathbbE[Verthat{f}_{a}lpha-f^{d}aggerVer{t}^2]$ w.r.t.~the true solution $f^{d}agger$ is minimized. In this paper, we introduce the Sharp Optimal Lepskiu{i}-Inspired Tuning (SOLIT) method, which yields an a posteriori parameter choice rule ensuring adaptive minimax rates of convergence. It depends only on $Y$ and the noise level $sigma$ as well as the operator $T$ and the filter $left(q_{a}lphaight{)}_{alphainmathcalA}$ and does not require any problem-dependent tuning of further parameters. We prove an oracle inequality for the corresponding MSE in a general setting and derive the rates of convergence in different scenarios. By a careful analysis we show that no other a posteriori parameter choice rule can yield a better performance in terms of the convergence rate of the MSE. In particular, our results reveal that the typical understanding of Lepskiii-type methods in inverse problems leading to a loss of a log factor is wrong. In addition, the empirical performance of SOLIT is examined in simulations.

Full work available at URL: https://arxiv.org/abs/2304.10356

zbMATH Keywords

adaptationoracle inequalityconvergence rateschoice of parametersordered filters

Mathematics Subject Classification ID

Linear operators and ill-posed problems, regularization (47A52)

Cites Work

• Introduction to nonparametric estimation
• A new chi-square approximation to the distribution of non-negative definite quadratic forms in non-central normal variables
• Composite quantile regression and the oracle model selection theory
• Estimating nuisance parameters in inverse problems
• The mathematics of computerized tomography
• Regularization of some linear ill-posed problems with discretized random noisy data
• Image deblurring with Poisson data: from cells to galaxies
• Geometry of linear ill-posed problems in variable Hilbert scales
• Title not available (Why is that?)
• How general are general source conditions?
• Comparing parameter choice methods for regularization of ill-posed problems
• Optimal filtering of square-integrable signals in Gaussian noise
• Randomized algorithms for estimating the trace of an implicit symmetric positive semi-definite matrix
• Minimax theory of image reconstruction
• Convergence Rates for Inverse Problems with Impulsive Noise
• Convergence rates in expectation for Tikhonov-type regularization of inverse problems with Poisson data
• On Difference-Based Variance Estimation in Nonparametric Regression When the Covariate is High Dimensional
• Statistical Inverse Estimation in Hilbert Scales
• Convergence Rates of General Regularization Methods for Statistical Inverse Problems and Applications
• The Lepskii principle revisited
• Regularization of exponentially ill-posed problems
• A Lepskij-type stopping rule for regularized Newton methods
• Regularization without preliminary knowledge of smoothness and error behaviour
• On the best rate of adaptive estimation in some inverse problems
• Title not available (Why is that?)
• A statistical approach to some inverse problems for partial differential equations
• Variance estimation for high-dimensional regression models
• Inverse problems with Poisson data: statistical regularization theory, applications and algorithms
• Convergence rates for exponentially ill-posed inverse problems with impulsive noise
• Adaptivity and Oracle Inequalities in Linear Statistical Inverse Problems: A (Numerical) Survey
• Inverse problems for partial differential equations
• Empirical risk minimization as parameter choice rule for general linear regularization methods
• Title not available (Why is that?)
• The Le Cam distance between density estimation, Poisson processes and Gaussian white noise
• Bootstrap tuning in Gaussian ordered model selection
• Minimax estimation of the solution of an ill-posed convolution type problem

Cited In (1)

• A posterior contraction for Bayesian inverse problems in Banach spaces

This page was built for publication: Adaptive minimax optimality in statistical inverse problems via SOLIT—Sharp Optimal Lepskiĭ-Inspired Tuning

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136775)