mlsanityData to 7 Oct 2026

WISE: full-Waveform variational Inference via Subsurface Extensions

arXiv 2401.06230 ↗First posted 11 Dec 2023physics.geo-ph cs.AI cs.LGcited by 10

Not on the map. Why?

Abstract (arXiv)

We introduce a probabilistic technique for full-waveform inversion, employing variational inference and conditional normalizing flows to quantify uncertainty in migration-velocity models and its impact on imaging. Our approach integrates generative artificial intelligence with physics-informed common-image gathers, reducing reliance on accurate initial velocity models. Considered case studies demonstrate its efficacy producing realizations of migration-velocity models conditioned by the data. These models are used to quantify amplitude and positioning effects during subsequent imaging.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 20 of its 62 references that are in the map. First posted on arXiv 11 Dec 2023.

Builds on

20 of 62 references.

References · 20

  1. 2022Hierarchical Text-Conditional Image Generation with CLIP Latentscited by 6,539
  2. 2015Variational Inference with Normalizing Flowscited by 2,161
  3. 2019Learning Likelihoods with Conditional Normalizing Flowscited by 170
  4. 2020BayesFlow: Learning complex stochastic models with invertible neural networkscited by 139
  5. 2018Iterative Amortized Inferencecited by 93
  6. 2021OpenFWI: Large-Scale Multi-Structural Benchmark Datasets for Seismic Full Waveform Inversioncited by 44
  7. 2019HINT: Hierarchical Invertible Neural Transport for Density Estimation and Bayesian Inferencecited by 31
  8. 2021Unsupervised Learning of Full-Waveform Inversion: Connecting CNN and Partial Differential Equation in a Loopcited by 29
  9. 2022Reliable amortized variational inference with physics-based latent distribution correctioncited by 25
  10. 2021Preconditioned training of normalizing flows for variational inference in inverse problemscited by 19
  11. 2021Deep Bayesian inference for seismic imaging with taskscited by 12
  12. 2020Parameterizing uncertainty by deep invertible networks, an application to reservoir characterizationcited by 11
8 more
  1. 2021Learning by example: fast reliability-aware seismic imaging with normalizing flowscited by 10
  2. 2023A Conditional Normalizing Flow for Accelerated Multi-Coil MR Imagingcited by 10
  3. 2023An Empirical Study of Large-Scale Data-Driven Full Waveform Inversioncited by 9
  4. 2023Amortized Normalizing Flows for Transcranial Ultrasound with Uncertainty Quantificationcited by 8
  5. 2023Learned multiphysics inversion with differentiable programming and machine learningcited by 7
  6. 2023Inference of CO2 flow patterns -- a feasibility studycited by 4
  7. 2023Refining Amortized Posterior Approximations using Gradient-Based Summary Statisticscited by 3
  8. 2022Velocity continuation with Fourier neural operators for accelerated uncertainty quantificationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 5; 2025: 1; 2026: 4.

10 papers.

  1. 2024Controllable seismic velocity synthesis using generative diffusion modelscited by 13
  2. 2024Paired Autoencoders for Likelihood-free Estimation in Inverse Problemscited by 6
  3. 2024Back-Projection Diffusion: Solving the Wideband Inverse Scattering Problem with Diffusion Modelscited by 3
  4. 2024Integrating Physics of the Problem into Data-Driven Methods to Enhance Elastic Full-Waveform Inversion with Uncertainty Quantificationcited by 1
  5. 2025Power-scaled Bayesian Inference with Score-based Generative Modelscited by 1
  6. 2026On the role of memorization in learned priors for geophysical inverse problemscited by 1
  7. 2026OpenSeisML: Open Large-Scale Real Seismic and well-log Dataset for Generative AIcited by 1
  8. 2024BEACON: Bayesian Experimental design Acceleration with Conditional Normalizing flows - a case study in optimal monitor well placement for CO2 sequestrationcited by 0
  9. 2026SAGE: Subsurface AI-driven Geostatistical Extraction with proxy posteriorcited by 0
  10. 2026Priors learned from legacy reconstructions inherit undetectable overconfidencecited by 0