Abstract:The Bayesian approach to solving inverse problems relies on the choice of a prior. This critical ingredient allows expert knowledge or physical constraints to be formulated in a probabilistic fashion and plays an important role for the success of the inference. Recently, Bayesian inverse problems were solved using generative models as highly informative priors. Generative models are a popular tool in machine learning to generate data whose properties closely resemble those of a given database. Typically, the generated distribution of data is embedded in a low-dimensional manifold. For the inverse problem, a generative model is trained on a database that reflects the properties of the sought solution, such as typical structures of the tissue in the human brain in magnetic resonance imaging. The inference is carried out in the low-dimensional manifold determined by the generative model that strongly reduces the dimensionality of the inverse problem. However, this procedure produces a posterior that does not admit a Lebesgue density in the actual variables and the accuracy attained can strongly depend on the quality of the generative model. For linear Gaussian models, we explore an alternative Bayesian inference based on probabilistic generative models; this inference is carried out in the original high-dimensional space. A Laplace approximation is employed to analytically derive the prior probability density function required, which is induced by the generative model. Properties of the resulting inference are investigated. Specifically, we show that derived Bayes estimates are consistent, in contrast to the approach in which the low-dimensional manifold of the generative model is employed. The MNIST data set is used to design numerical experiments that confirm our theoretical findings. It is shown that the approach proposed can be advantageous when the information contained in the data is high and a simple heuristic is considered for the detection of this case. Finally, the pros and cons of both approaches are discussed.

Consistency of Bayesian inference with Gaussian process priors in an elliptic inverse problem

Consistency of Bayesian inference with Gaussian process priors for a parabolic inverse problem

A Bayesian approach with Gaussian priors to the inverse problem of source identification in elliptic PDEs

Consistency of variational Bayesian inference for non-linear inverse problems of partial differential equations

Statistical guarantees for Bayesian uncertainty quantification in non-linear inverse problems with Gaussian process priors

A posterior contraction for Bayesian inverse problems in Banach spaces

Solving Geophysical Inversion Problems with Intractable Likelihoods: Linearized Gaussian Approximations Versus the Correlated Pseudo-marginal Method

Variational Gaussian Processes For Linear Inverse Problems

On the Frequentist Coverage of Bayes Posteriors in Nonlinear Inverse Problems

Convergence of Gaussian Process Regression with Estimated Hyper-Parameters and Applications in Bayesian Inverse Problems

On some information-theoretic aspects of non-linear statistical inverse problems

Linear methods for non-linear inverse problems

Introduction To Gaussian Process Regression In Bayesian Inverse Problems, With New ResultsOn Experimental Design For Weighted Error Measures

Bernstein - von Mises theorems for statistical inverse problems I: Schrödinger equation

Continuous Gaussian mixture solution for linear Bayesian inversion with application to Laplace priors

On the convergence of the Laplace approximation and noise-level-robustness of Laplace-based Monte Carlo methods for Bayesian inverse problems

Generative models and Bayesian inversion using Laplace approximation

Posterior Contraction for Empirical Bayesian Approach to Inverse Problems under Non-Diagonal Assumption

Taming Score-Based Diffusion Priors for Infinite-Dimensional Nonlinear Inverse Problems

Deep Gaussian Process Priors for Bayesian Image Reconstruction

Posterior Consistency for Gaussian Process Approximations of Bayesian Posterior Distributions