Adaptive Feedback Convolutional‐Neural‐Network‐Based High‐Resolution Reflection‐Waveform Inversion

Wu, Yulang; McMechan, George A.; Wang, Yanfei

doi:10.1029/2022jb024138

Cited by 18 publications

(7 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Y. Wu et al. (2022) design a dual loop framework for full waveform inversion using reflection data, where the inner loop trains a CNN to update the velocity model from the image, and the outer loop updates the training data set based on the results from the inner loop.…”

Section: Highlightsmentioning

confidence: 99%

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

O’Malley

Beroza

et al. 2023

JGR Solid Earth

View full text Add to dashboard Cite

After decades of low but continuing activity, applications of machine learning (ML) in solid Earth geoscience have exploded in popularity. This special collection provides a snapshot of those applications, which range from data processing to inversion and interpretation, for which ML appears particularly well suited. Inevitably, there are variations in the degree to which these methods have been developed. We hope that the progress seen in some areas will inspire efforts in others. Challenges remain, including the formidable task of how geoscience can keep pace with developments in ML while ensuring the scientific rigor that our field depends on, but with improvements in sensor technology and accelerating rates of data accumulation, the methods of ML seem poised to play an important role for the foreseeable future.

show abstract

Section: Highlightsmentioning

confidence: 99%

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

O’Malley

Beroza

et al. 2023

JGR Solid Earth

View full text Add to dashboard Cite

show abstract

“…( 9) depends on a good initial guess of the solution. A variety of approaches have been developed to address these issues [53], [54], [55], [56]. Specifically, Zhang and Gao [53], [54] developed a two-stage learning strategy by unrolling the FWI iteration.…”

Section: Physics Models Enhanced By Machine Learningmentioning

confidence: 99%

“…Specifically, Zhang and Gao [53], [54] developed a two-stage learning strategy by unrolling the FWI iteration. More recently, the unrolling technique was combined with forward modeling and prior knowledge [56]. The proposed approach involved nested iterations, with the inner loop corresponding to prediction of the velocity map from the training set and migration image, and the outer loop improving the training set using the velocity map estimated within the inner loop.…”

Section: Physics Models Enhanced By Machine Learningmentioning

confidence: 99%

“…Summary of machine learning based seismic inversion techniques addressed in this paper. of this paper, there are some open packages: InversionNet[7], VelocityGAN[41] 10 , and CNN-RWI11 [56].Both InversionNet and VelocityGAN are PyTorch-based packages for solving FWI based on acoustic wave equation. Extensive examples, tests, and benchmarks are provided for running both codes on OpenFWI datasets.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Physics-Guided Data-Driven Seismic Inversion: Recent Progress and Future Opportunities in Full Waveform Inversion

Lin

Theiler

Wohlberg

2022

Preprint

View full text Add to dashboard Cite

The goal of seismic inversion is to obtain subsurface properties from surface measurements. Seismic images have proven valuable, even crucial, for a variety of applications, including subsurface energy exploration, earthquake early warning, carbon capture and sequestration, estimating pathways of sub-surface contaminant transport, etc. These subsurface properties˜(such as wave speed, density, impedance, and reflectivity) influence the transmission of seismic waves through the subsurface media, and wellunderstood physics models (so-called "forward models") can be used to predict what surface measurements would be made, for any given subsurface configuration. Seismic inversion is the inverse problem: given actual surface measurements, infer what

show abstract

“…Y. Wu et al. (2022) proposed a CNN‐based reflection‐waveform inversion method to iteratively update the CNN to predict the model. However, most of the aforementioned methods depend on supervised learning, which requires a large representative training dataset composed of seismic data and corresponding true velocity models.…”

Section: Introductionmentioning

confidence: 99%

FWIGAN: Full‐Waveform Inversion via a Physics‐Informed Generative Adversarial Network

Yang

2023

JGR Solid Earth

View full text Add to dashboard Cite

Full‐waveform inversion (FWI) is a powerful geophysical imaging technique that reproduces high‐resolution subsurface physical parameters by iteratively minimizing the misfit between the simulated and observed seismograms. Unfortunately, conventional FWI with a least‐squares loss function suffers from various drawbacks, such as the local‐minima problem and human intervention in the fine‐tuning of parameters. It is particular problematic when applied with noisy data and inadequate starting models. Recent work relying on partial differential equations and neural networks show promising performance in two‐dimensional FWI. Inspired by the competitive learning of generative adversarial networks, we propose an unsupervised learning paradigm that integrates the wave equation with a discriminative network to accurately estimate physically consistent velocity models in a distributional sense (FWIGAN). The introduced framework does not require a labeled training dataset or pretraining of the network; therefore, this framework is flexible and able to achieve inversion with minimal user interaction. We experimentally validate our method for three baseline geological models, and a comparison of the results demonstrates that FWIGAN faithfully recovers the velocity models and consistently outperforms other traditional or deep learning‐based algorithms. A further benefit from the physics‐constrained learning used in this method is that FWIGAN mitigates the local‐minima issue by reducing the sensitivity to initial models or data noise.

show abstract

Adaptive Feedback Convolutional‐Neural‐Network‐Based High‐Resolution Reflection‐Waveform Inversion

Cited by 18 publications

References 58 publications

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

Machine Learning Developments and Applications in Solid‐Earth Geosciences: Fad or Future?

Physics-Guided Data-Driven Seismic Inversion: Recent Progress and Future Opportunities in Full Waveform Inversion

FWIGAN: Full‐Waveform Inversion via a Physics‐Informed Generative Adversarial Network

Contact Info

Product

Resources

About