Random forward models and log-likelihoods in Bayesian inverse problems
From MaRDI portal
Abstract: We consider the use of randomised forward models and log-likelihoods within the Bayesian approach to inverse problems. Such random approximations to the exact forward model or log-likelihood arise naturally when a computationally expensive model is approximated using a cheaper stochastic surrogate, as in Gaussian process emulation (kriging), or in the field of probabilistic numerical methods. We show that the Hellinger distance between the exact and approximate Bayesian posteriors is bounded by moments of the difference between the true and approximate log-likelihoods. Example applications of these stability results are given for randomised misfit models in large data applications and the probabilistic solution of ordinary differential equations.
Recommendations
- Approximation of Bayesian Inverse Problems for PDEs
- Adaptive construction of surrogates for the Bayesian solution of inverse problems
- Error bounds for some approximate posterior measures in Bayesian inference
- Solution of inverse problems with limited forward solver evaluations: a Bayesian perspective
- A Randomized Maximum A Posteriori Method for Posterior Sampling of High Dimensional Nonlinear Bayesian Inverse Problems
Cites work
- A data-scalable randomized misfit approach for solving large-scale PDE-constrained inverse problems
- A stability property of implicit Runge-Kutta methods
- Bayesian solution uncertainty quantification for differential equations
- Besov priors for Bayesian inverse problems
- Brittleness of Bayesian inference under finite information in a continuous world
- Can one use total variation prior for edge-preserving Bayesian inversion?
- Data Assimilation
- Displacement convexity of generalized relative entropies
- scientific article; zbMATH DE number 3112287 (Why is no real title available?)
- scientific article; zbMATH DE number 5688205 (Why is no real title available?)
- scientific article; zbMATH DE number 4098703 (Why is no real title available?)
- scientific article; zbMATH DE number 1350773 (Why is no real title available?)
- scientific article; zbMATH DE number 1153603 (Why is no real title available?)
- scientific article; zbMATH DE number 802686 (Why is no real title available?)
- scientific article; zbMATH DE number 862361 (Why is no real title available?)
- scientific article; zbMATH DE number 3202900 (Why is no real title available?)
- Inverse problems as statistics
- Inverse problems: a Bayesian perspective
- Lectures on Stochastic Programming
- MCMC methods for functions: modifying old algorithms to make them faster
- Measure theory. Vol. I and II
- Posterior consistency for Gaussian process approximations of Bayesian posterior distributions
- Probabilistic Forecasting and Bayesian Data Assimilation
- Probabilistic numerics and uncertainty in computations
- Qualitative robustness in Bayesian inference
- Robust Stochastic Approximation Approach to Stochastic Programming
- Statistical analysis of differential equations: introducing probability measures on numerical solutions
- Statistical and computational inverse problems.
- Strong Convergence of Euler-Type Methods for Nonlinear Stochastic Differential Equations
- Strong convergence rates of probabilistic integrators for ordinary differential equations
- The pseudo-marginal approach for efficient Monte Carlo computations
- Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors
- Well-posed Bayesian inverse problems with infinitely divisible and heavy-tailed prior measures
- Well-Posed Bayesian Inverse Problems: Priors with Exponential Tails
Cited in
(17)- Randomized reduced forward models for efficient Metropolis-Hastings MCMC, with application to subsurface fluid flow and capacitance tomography
- A role for symmetry in the Bayesian solution of differential equations
- Randomised one-step time integration methods for deterministic operator differential equations
- Randomized approaches to accelerate MCMC algorithms for Bayesian inverse problems
- Random time step probabilistic methods for uncertainty quantification in chaotic and geometric numerical integration
- Stability of doubly-intractable distributions
- A probabilistic finite element method based on random meshes: a posteriori error estimators and Bayesian inverse problems
- Strong convergence rates of probabilistic integrators for ordinary differential equations
- A modern retrospective on probabilistic numerics
- Inverse Analysis of a New Anomalous Diffusion Model Employing Maximum Likelihood and Bayesian Estimation
- Statistical treatment of inverse problems constrained by differential equations-based models with stochastic terms
- Convergence of Gaussian process regression with estimated hyper-parameters and applications in Bayesian inverse problems
- Error bounds for some approximate posterior measures in Bayesian inference
- On the importance of the Jacobian determinant in parameter inference for random parameter and random measurement error models
- Error control of the numerical posterior with Bayes factors in Bayesian uncertainty quantification
- Gaussian processes for Bayesian inverse problems associated with linear partial differential equations
- Introduction to Gaussian process regression in Bayesian inverse problems, with new results on experimental design for weighted error measures
This page was built for publication: Random forward models and log-likelihoods in Bayesian inverse problems
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4611529)
