PDE-based Bayesian Inverse Problems in All-at-once Framework

DOI10.48550/ARXIV.2610.39783MaRDI QIDQ7409110arXivFDO

Authors Fabian Schneider, Tram Thi Ngoc Nguyen, Leila Taghizadeh

Publication date 29 October 2026

Published in arXiv

Tldr We propose an efficient PDE-constrained Bayesian inversion framework by modeling the PDE residual as probabilistic information and eliminating the need for repeated PDE solves in each iteration. The resulting Bayesian all-at-once approach yields a significant speedup (factor of two) while maintaining comparable reconstruction quality compared to traditional reduced-order methods.

Full work available at URL https://doi.org/10.48550/ARXIV.2610.39783

arXiv classification

math.NAmath.PR

Cites work

• A computational framework for infinite-dimensional Bayesian inverse problems. I: The linearized case, with application to global seismic inversion
• A Convex Approximation Framework for Neural Likelihood-Based Bayesian Inverse Problems
• A penalty method for PDE-constrained optimization in inverse problems
• A reduced SAND method for optimal design of non-linear structures
• Airfoil Design by an All-at-once Method*
• All-at-once formulation meets the Bayesian approach: a study of two prototypical linear inverse problems
• All-at-once versus reduced iterative methods for time dependent inverse problems
• An introduction to the mathematical theory of inverse problems
• Bayesian inversion for the identification of the doping profile in unipolar semiconductor devices
• Bi-level iterative regularization for inverse problems in nonlinear PDEs
• Deep hidden physics models: deep learning of nonlinear partial differential equations
• Discretization of parameter identification in PDEs using Neural Networks
• Global optimality conditions for sensor placement, with extensions to binary A-optimal experimental designs
• Goal oriented adaptivity in the IRGNM for parameter identification in PDEs. II: All-at-once formulations
• Identification of doping profiles in semiconductor devices
• Inexact SQP Interior Point Methods and Large Scale Optimal Control Problems
• Inverse problems: a Bayesian perspective
• Iterative regularization of parameter identification problems by sequential quadratic programming methods
• Landweber-Kaczmarz for parameter identification in time-dependent inverse problems: All-at-once versus Reduced version
• Learning-informed parameter identification in nonlinear time-dependent PDEs
• Multilevel dimension-independent likelihood-informed MCMC for large-scale inverse problems
• Numerical Approximation of an SQP-Type Method for Parameter Identification
• Numerical solution of a nonlinear parabolic control problem by a reduced SQP method
• On the convergence analysis of one-shot inversion methods
• On the convergence analysis of one-shot inversion methods
• Parameter identification for the Landau-Lifshitz-Gilbert equation in magnetic particle imaging
• Partial differential equations
• Physics-informed neural networks: a deep learning framework for solving forward and inverse problems involving nonlinear partial differential equations
• Preconditioned all-at-once methods for large, sparse parameter estimation problems
• Reduced SQP Methods for Parameter Identification Problems
• Regularization based on all-at-once formulations for inverse problems
• Regularization of Inverse Problems
• Sequential bi-level regularized inversion with application to hidden reaction law discovery
• The Bayesian Approach To Inverse Problems
• The extended adjoint state and nonlinearity in correlation-based passive imaging
• The extended adjoint state and nonlinearity in correlation-based passive imaging
• The tangential cone condition for some coefficient identification model problems in parabolic PDEs
• Uncertainty Quantification and Weak Approximation of an Elliptic Inverse Problem

This page was built for publication: PDE-based Bayesian Inverse Problems in All-at-once Framework

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q7409110)