Efficient nonparametric Bayesian inference for X-ray transforms

DOI10.1214/18-AOS1708MaRDI QIDQ666595zbMATH OpenOpenAlexFDO

Authors François Monard, Richard Nickl, Gabriel P. Paternain

Publication date 6 March 2019

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

Full work available at URL https://arxiv.org/abs/1708.06332, https://projecteuclid.org/euclid.aos/1547197250

zbMATH Keywords

reproducing kernel Hilbert spaceGaussian processinverse problemsemiparametric efficiencymicrolocal analysisRadon transformGaussian priorMAP estimateBernstein-von Mises theorem\(X\)-rayTikhonov regulariser

Mathematics Subject Classification ID

Bayesian inference (62F15) Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Applications of statistics to physics (62P35) Radon transform (44A12)

Abstract: We consider the statistical inverse problem of recovering a function $f:MomathbbR$, where $M$ is a smooth compact Riemannian manifold with boundary, from measurements of general $X$-ray transforms $I_a(f)$ of $f$, corrupted by additive Gaussian noise. For $M$ equal to the unit disk with `flat' geometry and $a=0$ this reduces to the standard Radon transform, but our general setting allows for anisotropic media $M$ and can further model local `attenuation' effects -- both highly relevant in practical imaging problems such as SPECT tomography. We propose a nonparametric Bayesian inference approach based on standard Gaussian process priors for $f$. The posterior reconstruction of $f$ corresponds to a Tikhonov regulariser with a reproducing kernel Hilbert space norm penalty that does not require the calculation of the singular value decomposition of the forward operator $I_a$. We prove Bernstein-von Mises theorems that entail that posterior-based inferences such as credible sets are valid and optimal from a frequentist point of view for a large family of semi-parametric aspects of $f$. In particular we derive the asymptotic distribution of smooth linear functionals of the Tikhonov regulariser, which is shown to attain the semi-parametric Cram'er-Rao information bound. The proofs rely on an invertibility result for the `Fisher information' operator $I_a^{\ast }{I}_a$ between suitable function spaces, a result of independent interest that relies on techniques from microlocal analysis. We illustrate the performance of the proposed method via simulations in various settings.

Recommendations

• Statistical guarantees for Bayesian uncertainty quantification in nonlinear inverse problems with Gaussian process priors
• Bayesian inverse problems with non-conjugate priors
• Bernstein-von Mises theorems and uncertainty quantification for linear inverse problems
• Bayesian inverse problems with Gaussian priors
• Non-Gaussian statistical inverse problems. Part I: Posterior distributions

Cites work

• scientific article; zbMATH DE number 3969124 (Why is no real title available?)
• scientific article; zbMATH DE number 4001918 (Why is no real title available?)
• scientific article; zbMATH DE number 3709066 (Why is no real title available?)
• scientific article; zbMATH DE number 3766258 (Why is no real title available?)
• scientific article; zbMATH DE number 3567529 (Why is no real title available?)
• scientific article; zbMATH DE number 2151809 (Why is no real title available?)
• scientific article; zbMATH DE number 914800 (Why is no real title available?)
• scientific article; zbMATH DE number 1446717 (Why is no real title available?)
• scientific article; zbMATH DE number 3273551 (Why is no real title available?)
• scientific article; zbMATH DE number 2239537 (Why is no real title available?)
• A Bernstein-von Mises theorem for smooth functionals in semiparametric models
• Adaptive estimation of linear functionals in Hilbert scales from indirect white noise observa\-tions
• An inversion formula for the attenuated \(X\)-ray transformation
• Asymptotic Statistics
• Bayesian inverse problems with Gaussian priors
• Bayesian inverse problems with non-conjugate priors
• Bernstein-von Mises theorems for statistical inverse problems. I: Schrödinger equation
• Boundary problems for pseudo-differential operators
• Closed geodesics on non-compact Riemannian manifolds
• Comportement d'un opérateur pseudo-différentiel sur une variété à bord. I: La propriété de transmission, II: Pseudo-noyaux de Poisson
• Fractional Laplacians on domains, a development of Hörmander's theory of \(\mu\)-transmission pseudodifferential operators
• Fundamentals of nonparametric Bayesian inference
• Integral geometry for tensor fields. Transl. from the Russian
• Inverse problems: a Bayesian perspective
• Inversion of noisy Radon transform by SVD based needlets
• Local and nonlocal boundary conditions for \(\mu\)-transmission and fractional elliptic pseudodifferential operators
• Low-rank independence samplers in hierarchical Bayesian inverse problems
• MAP estimators and their consistency in Bayesian nonparametric inverse problems
• Mathematical foundations of infinite-dimensional statistical models
• Nonparametric Bayesian posterior contraction rates for discretely observed scalar diffusions
• Nonparametric Bernstein-von Mises theorems in Gaussian white noise
• Nonparametric statistical inverse problems
• Numerical implementation of geodesic X-ray transforms and their inversion
• On Bayesian supremum norm contraction rates
• On the Bernstein-von Mises phenomenon for nonparametric Bayes procedures
• Oracle inequalities for inverse problems
• Orthogonal Function Series Expansions and the Null Space of the Radon Transform
• Pólya tree posterior distributions on densities
• Rates of contraction of posterior distributions based on Gaussian process priors
• Real Analysis and Probability
• Speed of estimation in positron emission tomography and related inverse problems
• Stability estimates for the \(X\)-ray transform of tensor fields and boundary rigidity
• The Radon transform and medical imaging
• The Radon transform.
• The X-ray transform for a generic family of curves and weights
• The attenuated ray transform on simple surfaces
• The boundary rigidity problem in the presence of a magnetic field
• The geodesic X-ray transform with matrix weights
• The inverse problem for the local geodesic ray transform
• Two dimensional compact simple Riemannian manifolds are boundary distance rigid

Cited in

(35)

• Bernstein-von Mises theorems for statistical inverse problems. I: Schrödinger equation
• Sharp stability estimate for the geodesic ray transform
• Non-asymptotic error estimates for the Laplace approximation in Bayesian inverse problems
• On statistical Calderón problems
• Double b-fibrations and desingularization of the X-ray transform on manifolds with strictly convex boundary
• Consistency of Bayesian inference with Gaussian process priors in an elliptic inverse problem
• Functional relations, sharp mapping properties, and regularization of the X-ray transform on disks of constant curvature
• Nonparametric statistical inference for drift vector fields of multi-dimensional diffusions
• Bernstein-von Mises theorems for statistical inverse problems. II: Compound Poisson processes
• Minimax optimal estimator in a stochastic inverse problem for exponential Radon transform
• A Bernstein–von-Mises theorem for the Calderón problem with piecewise constant conductivities
• On the mean first arrival time of Brownian particles on Riemannian manifolds
• Functions of constant geodesic x-ray transform
• Dimension Free Nonasymptotic Bounds on the Accuracy of High-Dimensional Laplace Approximation
• Reconstruction of piecewise constant functions from x-ray data
• A sharp stability estimate for tensor tomography in non-positive curvature
• On the Bernstein-von Mises theorem for the Dirichlet process
• Convergence Rates for Penalized Least Squares Estimators in PDE Constrained Regression Problems
• Edge-promoting adaptive Bayesian experimental design for X-ray imaging
• Consistency of Bayesian inference with Gaussian process priors for a parabolic inverse problem
• Gaussian process regression in the flat limit
• On some information-theoretic aspects of non-linear statistical inverse problems
• Adaptive estimation of a function from its exponential Radon transform in presence of noise
• Identification of nonlinear beam-hardening effects in X-ray tomography
• Optimization based methods for partially observed chaotic systems
• Bernstein-von Mises theorems and uncertainty quantification for linear inverse problems
• Statistical guarantees for Bayesian uncertainty quantification in nonlinear inverse problems with Gaussian process priors
• Re-thinking high-dimensional mathematical statistics. Abstracts from the workshop held May 15--21, 2022
• On log-concave approximations of high-dimensional posterior measures and stability properties in non-linear inverse problems
• Bayesian linear inverse problems in regularity scales
• Convergence Rates for Learning Linear Operators from Noisy Data
• The C∞ -isomorphism property for a class of singularly-weighted x-ray transforms
• Solving inverse problems using data-driven models
• On empirical Bayes approach to inverse problems
• Consistency of the Bayes method for the inverse scattering problem

This page was built for publication: Efficient nonparametric Bayesian inference for \(X\)-ray transforms

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q666595)