Geophysical inversion versus machine learning in inverse problems

Геофизическая инверсия в сравнении с машинным обучением в обратных задачах
Yuji Kim, Nori Nakata
2018-12-01

L1 regularizationartificial neural networksgeophysical inversionmachine learningreflectivity inversion
Abstract Geophysical inversion and machine learning both provide solutions for inverse problems in which we estimate model parameters from observations. Geophysical inversions such as impedance inversion, amplitude-variation-with-offset inversion, and traveltime tomography are commonly used in the industry to yield physical properties from measured seismic data. Machine learning, a data-driven approach, has become popular during the past decades and is useful for solving such inverse problems. An advantage of machine learning methods is that they can be implemented without knowledge of physical equations or theories. The challenges of machine learning lie in acquiring enough training data and selecting relevant parameters, which are essential in obtaining a good quality model. In this study, we compare geophysical inversion and machine learning approaches in solving inverse problems and show similarities and differences of these approaches in a mathematical form and numerical tests. Both methods aid in solving ill-posed and nonlinear problems and use similar optimization techniques. We take reflectivity inversion as an example of the inverse problem. We apply geophysical inversion based on the least-squares method and artificial neural networks as a machine learning approach to solve reflectivity inversion using 2D synthetic data sets and 3D field data sets. A neural network with multiple hidden layers successfully generates the nonlinear mapping function to predict reflectivity. For this inverse problem, we test different L1 regularizers for both approaches. L1 regularization alleviates effects of noise in seismic traces and enhances sparsity, especially in the least-squares method. The 2D synthetic wedge model and field data examples show that neural networks yield high spatial resolution.
1
Geophysical inversion and machine learning address ill-posed, nonlinear inverse problems using mathematically similar optimization techniques.
2
In reflectivity inversion, multilayer neural networks successfully learn nonlinear mappings from seismic observations to predicted reflectivity.
3
L1 regularization reduces noise effects and promotes sparsity in both approaches, with particularly strong benefits for least-squares inversion.
4
Machine learning avoids requiring explicit physical equations but depends critically on sufficient training data and appropriate parameter selection.
5
Synthetic wedge and field-data experiments show that neural networks produce high-spatial-resolution reflectivity estimates.

Reflectivity inversion from seismic data using geophysical inversion and machine learning approaches

Comparison of inversion accuracy, noise robustness, sparsity enhancement, and spatial resolution between least-squares geophysical inversion and artificial neural networks, including the effects of L1 regularization

Publication Details
Publication Date
2018-12-01
Journal
Publisher
ISSN
Cited by
162
Access Type
Author Information
Authors
Yuji Kim
Nori Nakata
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%