Stability of Gibbs posteriors from the Wasserstein loss for Bayesian full waveform inversion
From MaRDI portal
Publication:5158929
Abstract: Recently, the Wasserstein loss function has been proven to be effective when applied to deterministic full-waveform inversion (FWI) problems. We consider the application of this loss function in Bayesian FWI so that the uncertainty can be captured in the solution. Other loss functions that are commonly used in practice are also considered for comparison. Existence and stability of the resulting Gibbs posteriors are shown on function space under weak assumptions on the prior and model. In particular, the distribution arising from the Wasserstein loss is shown to be quite stable with respect to high-frequency noise in the data. We then illustrate the difference between the resulting distributions numerically, using Laplace approximations to estimate the unknown velocity field and uncertainty associated with the estimates.
Recommendations
Cites work
- scientific article; zbMATH DE number 1909499 (Why is no real title available?)
- A Bayesian level set method for geometric inverse problems
- A new class of transport distances between measures
- Absorbing boundary conditions for numerical simulation of waves
- An optimal transport approach for seismic tomography: application to 3D full waveform inversion
- Application of the Wasserstein metric to seismic signals
- Bayesian inference with optimal maps
- Bayesian inversion in resin transfer molding
- Calibrating general posterior credible regions
- Comparison between \(W_2\) distance and \(\dot{H}^{-1}\) norm, and localization of Wasserstein distance
- Generalization of an inequality by Talagrand and links with the logarithmic Sobolev inequality
- Geometric MCMC for infinite-dimensional inverse problems
- Hierarchical Bayesian level set inversion
- Inference for empirical Wasserstein distances on finite spaces
- Inverse Problem Theory and Methods for Model Parameter Estimation
- On the convergence of the Laplace approximation and noise-level-robustness of Laplace-based Monte Carlo methods for Bayesian inverse problems
- On the properties of variational approximations of Gibbs posteriors
- On the well-posedness of Bayesian inverse problems
- Optimal transport for seismic full waveform inversion
- Reconciling Bayesian and perimeter regularization for binary inversion
- Scalable and efficient algorithms for the propagation of uncertainty from data through inference to prediction for large-scale problems, with application to flow of the antarctic ice sheet
- The quadratic Wasserstein metric for inverse data matching
- Uncertainty quantification in graph-based classification of high dimensional data
- WASSERSTEIN METRIC-DRIVEN BAYESIAN INVERSION WITH APPLICATIONS TO SIGNAL PROCESSING
- Well-posed Bayesian inverse problems and heavy-tailed stable quasi-Banach space priors
Cited in
(6)- Implicit regularization effects of the Sobolev norms in image processing
- Generalized Bayes approach to inverse problems with model misspecification
- Optimal Transport for Parameter Identification of Chaotic Dynamics via Invariant Measures
- Norm-dependent convergence and stability of the inverse scattering series for diffuse and scalar waves
- Object based Bayesian full-waveform inversion for shear elastography
- Applications of optimal transportation in the natural sciences. Abstracts from the workshop held February 21--27, 2021 (online meeting)
This page was built for publication: Stability of Gibbs posteriors from the Wasserstein loss for Bayesian full waveform inversion
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5158929)
