VIP - Variational Inversion Package with example implementations of Bayesian tomographic imaging

2024/05/29 by Xin Zhang, Andrew Curtis · 1 voice
Earth and Planetary Sciences · Engineering · Medicine · #Algorithm #Artificial intelligence #Bayesian probability #Computational science #Computer science #Drilling and Well Engineering #Geology #Implementation #Inversion (geology) #Iterative reconstruction #Medical Imaging Techniques and Applications #Paleontology #Programming language #R package #Seismic Imaging and Inversion Techniques #Tomographic reconstruction

paper · pdf · doi:10.26443/seismica.v3i1.1143

openalex publication_date 2024/05/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Bayesian inference has become an important methodology to solve inverse problems and to quantify uncertainties in their solutions. Variational inference is a method that provides probabilistic, Bayesian solutions efficiently by using optimisation. In this study we present a Python Variational Inversion Package (VIP), to solve inverse problems using variational inference methods. The package includes automatic differential variational inference (ADVI), Stein variational gradient descent (SVGD) and stochastic SVGD (sSVGD), and provides implementations of 2D travel time tomography and 2D full waveform inversion including test examples and solutions. Users can solve their own problems by supplying an appropriate forward function and a gradient calculation code. In addition, the package provides a scalable implementation which can be deployed easily on a desktop machine or using modern high performance computational facilities. The examples demonstrate that VIP is an efficient, scalable, extensible and user-friendly package, and can be used to solve a wide range of low or high dimensional inverse problems in practice.

Citations

• ADADELTA: An Adaptive Learning Rate Method
• On Information and Sufficiency
• A Stochastic Approximation Method
• Inversion of seismic reflection data in the acoustic approximation
• The adjoint method in seismology
• Monte Carlo sampling methods using Markov chains and their applications
• Variational Inference: A Review for Statisticians
• Automatic Differentiation Variational Inference
• pyGIMLi: An open-source library for modelling and inversion in geophysics
• Stein Variational Gradient Descent With Matrix-Valued Kernels
• Geopsy: A User-Friendly Open-Source Tool Set for Ambient Vibration Processing
• Faster Uncertainty Quantification for Inverse Problems with Conditional Normalizing Flows
• MCMC using Hamiltonian dynamics
• Videovigilancia de altura
• Monte Carlo sampling of solutions to inverse problems
• A review of the adjoint-state method for computing the gradient of a functional with geophysical applications
• Amortized Normalizing Flows for Transcranial Ultrasound with Uncertainty Quantification

Discussions

• Want a Bayesian inversion framework?  Try this new Variational Inversion Package (VIP) by Zhang & Curtis, with three variational methods and the ability to solve various problems, ranging from simple [bsky, 0 points, 0 comments]

Related

• VIP -- Variational Inversion Package with example implementations of Bayesian tomographic imaging
• Bayesian Inversion, Uncertainty Analysis and Interrogation using Boosting Variational Inference
• A variational inference framework for inverse problems
• ASPIRE: iterative amortized posterior inference for Bayesian inverse problems
• A probabilistic approach to tomography and adjoint state methods, with an application to full waveform inversion in medical ultrasound