Bingbing Sun
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Bingbing Sun has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
- Tariq Alkhalifah
Division of Physical Science and Engineering King Abdullah University of Science and Technology 23955-6900 Thuwal Saudi Arabia
15 shared publications - Abdullah Alali
Division of Physical Sciences and Engineering King Abdullah University of Science and Technology‐ (KAUST) Thuwal Saudi Arabia
2 shared publications Tariq Alkhalifah
Physical Science and Engineering, King Abdullah University of Science and Technology, Thuwal, SAUDI ARABIA
1 shared publication- Vladimir Kazei
Aramco Americas, Houston Research Center, Houston, Texas, USA.
1 shared publication - Yongjie Tang
Department of Earth and Space Sciences, Southern University of Science and Technology, Shenzhen 518055, China
1 shared publication - Zedong Wu
King Abdullah University of Science and Technology (KAUST), Physical Sciences and Engineering Division, Thuwal 23955, Saudi Arabia.
1 shared publication
Latest Publications
- 27 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 61 (01962892) , 1-22
An Anisotropic Waveform Inversion Using an Optimal Transport Matching Filter Objective: An Application to an Offshore Field Dataset
Considering the conventional seismic wavelength, and the nature of the Earth layering, seismic waves experience, in many parts of the subsurface, considerable anisotropy, and with the effect of gravity on sedimentation, the anisotropy tends to be of a transversely isotropic with a vertical axis of symmetry (VTI) nature. Inverting for such a model of the Earth using waveforms, we face considerable nonlinearity and parameter trade-off. A recently introduced optimal transport of the matching filter (OTMF) provided us with a robust misfit function for reducing cycle-skipping in Full-Waveform Inversion (FWI). We apply a VTI FWI using the OTMF misfit on a field dataset from offshore Australia, comparing its performance to that of conventional FWI using the L2-norm misfit in a variety of circumstances. Due to strong anisotropy in this region, an isotropic inversion though can fit the record, leading to common image gathers with sizable linear moveouts. Thus, in an anisotropic VTI setup, starting the inversion from 3 Hz, both the L2 norm and the OTMF misfit functions can generate a geologically meaningful model, and recover similar anisotropy anomalies. We demonstrate that the OTMF misfit, in some sense, can address the nonlinearity of FWI due to its intrinsic global updating features. Compared to the results from isotropic full-waveform inversion, the improvements in the RTM image and the common image gathers further demonstrate the benefits of including anisotropy in the FWI inversion engine, as well as, the good performance of the OTMF in mitigating cycle-skipping.
- 27 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 61 (01962892) , 1-22
An Anisotropic Waveform Inversion Using an Optimal Transport Matching Filter Objective: An Application to an Offshore Field Dataset
Considering the conventional seismic wavelength, and the nature of the Earth layering, seismic waves experience, in many parts of the subsurface, considerable anisotropy, and with the effect of gravity on sedimentation, the anisotropy tends to be of a transversely isotropic with a vertical axis of symmetry (VTI) nature. Inverting for such a model of the Earth using waveforms, we face considerable nonlinearity and parameter trade-off. A recently introduced optimal transport of the matching filter (OTMF) provided us with a robust misfit function for reducing cycle-skipping in Full-Waveform Inversion (FWI). We apply a VTI FWI using the OTMF misfit on a field dataset from offshore Australia, comparing its performance to that of conventional FWI using the L2-norm misfit in a variety of circumstances. Due to strong anisotropy in this region, an isotropic inversion though can fit the record, leading to common image gathers with sizable linear moveouts. Thus, in an anisotropic VTI setup, starting the inversion from 3 Hz, both the L2 norm and the OTMF misfit functions can generate a geologically meaningful model, and recover similar anisotropy anomalies. We demonstrate that the OTMF misfit, in some sense, can address the nonlinearity of FWI due to its intrinsic global updating features. Compared to the results from isotropic full-waveform inversion, the improvements in the RTM image and the common image gathers further demonstrate the benefits of including anisotropy in the FWI inversion engine, as well as, the good performance of the OTMF in mitigating cycle-skipping.
- Open Access
- 29 September 2022
- Published by Frontiers Media SA in Frontiers in Earth Science
- Vol. 10
ML-misfit: A neural network formulation of the misfit function for full-waveform inversion
A robust misfit function is essential for mitigating cycle-skipping in full-waveform inversion (FWI), leading to stable updates of the velocity model in this highly nonlinear optimization process. State-of-the-art misfit functions, including matching filter or optimal transport misfits, are all hand-crafted and developed from first principles. With the growth of artificial intelligence in geoscience, we propose learning a robust misfit function for FWI, entitled ML-misfit, based on machine learning. Inspired by the recently introduced optimal transport of the matching filter objective function, we design a specific neural network architecture for the misfit function in a form that allows for global comparison of the predicted and measured data. The proposed neural network architecture also guarantees that the resulting misfit is a pseudo-metric for efficient training. In the framework of meta-learning, we train the network by running FWI to invert for randomly generated velocity models and update the parameters of the neural network by minimizing the meta-loss, which is defined as the accumulated difference between the true and inverted velocity models. The learning and improvement of such an ML-misfit are automatic, and the resulting ML-misfit is data-adaptive. We first illustrate the basic principles behind the ML-misfit for learning a convex misfit function using a travel-time shifted signal example. Furthermore, we train the neural network on 2D horizontally layered models and apply the trained neural network to the Marmousi model; the resulting ML-misfit provides robust updating of the model and mitigates the cycle-skipping issue successfully.
- 29 September 2022
- Published by Frontiers Media SA in Frontiers in Earth Science
- Vol. 10
Open AccessML-misfit: A neural network formulation of the misfit function for full-waveform inversion
A robust misfit function is essential for mitigating cycle-skipping in full-waveform inversion (FWI), leading to stable updates of the velocity model in this highly nonlinear optimization process. State-of-the-art misfit functions, including matching filter or optimal transport misfits, are all hand-crafted and developed from first principles. With the growth of artificial intelligence in geoscience, we propose learning a robust misfit function for FWI, entitled ML-misfit, based on machine learning. Inspired by the recently introduced optimal transport of the matching filter objective function, we design a specific neural network architecture for the misfit function in a form that allows for global comparison of the predicted and measured data. The proposed neural network architecture also guarantees that the resulting misfit is a pseudo-metric for efficient training. In the framework of meta-learning, we train the network by running FWI to invert for randomly generated velocity models and update the parameters of the neural network by minimizing the meta-loss, which is defined as the accumulated difference between the true and inverted velocity models. The learning and improvement of such an ML-misfit are automatic, and the resulting ML-misfit is data-adaptive. We first illustrate the basic principles behind the ML-misfit for learning a convex misfit function using a travel-time shifted signal example. Furthermore, we train the neural network on 2D horizontally layered models and apply the trained neural network to the Marmousi model; the resulting ML-misfit provides robust updating of the model and mitigates the cycle-skipping issue successfully.
- 18 May 2022
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 87 (5) , V405-V417
Time-lapse data matching using a recurrent neural network approach
Time-lapse seismic data acquisition is an essential tool to monitor changes in a reservoir due to fluid injection, such as CO2 injection. By acquiring multiple seismic surveys in the exact same location, the authors can identify the reservoir changes by analyzing the difference in the data. However, such analysis can be skewed by the near-surface seasonal velocity variations, inaccuracy, and repeatability in the acquisition parameters, and other inevitable noise. The common practice (cross equalization) to address this problem uses the part of the data in which changes are not expected to design a matching filter and then apply it to the whole data, including the reservoir area. Like cross equalization, the authors train a recurrent neural network (RNN) on parts of the data excluding the reservoir area and then infer the reservoir-related data. The RNN can learn the time dependency of the data, unlike the matching filter that processes the data based on the local information obtained in the filter window. The authors determine the method of matching the data in various examples and compare it with the conventional matching filter. Specifically, we start by demonstrating the ability of the approach in matching two traces and then test the method on a prestack 2D synthetic data. Then, the authors verify the enhancements of the 4D signal by providing reverse time migration images. The authors measure the repeatability using normalized root-mean-square and predictability metrics and find that, in some cases, our proposed method performed better than the matching filter approach.
- 18 May 2022
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 87 (5) , V405-V417
Time-lapse data matching using a recurrent neural network approach
Time-lapse seismic data acquisition is an essential tool to monitor changes in a reservoir due to fluid injection, such as CO2 injection. By acquiring multiple seismic surveys in the exact same location, the authors can identify the reservoir changes by analyzing the difference in the data. However, such analysis can be skewed by the near-surface seasonal velocity variations, inaccuracy, and repeatability in the acquisition parameters, and other inevitable noise. The common practice (cross equalization) to address this problem uses the part of the data in which changes are not expected to design a matching filter and then apply it to the whole data, including the reservoir area. Like cross equalization, the authors train a recurrent neural network (RNN) on parts of the data excluding the reservoir area and then infer the reservoir-related data. The RNN can learn the time dependency of the data, unlike the matching filter that processes the data based on the local information obtained in the filter window. The authors determine the method of matching the data in various examples and compare it with the conventional matching filter. Specifically, we start by demonstrating the ability of the approach in matching two traces and then test the method on a prestack 2D synthetic data. Then, the authors verify the enhancements of the 4D signal by providing reverse time migration images. The authors measure the repeatability using normalized root-mean-square and predictability metrics and find that, in some cases, our proposed method performed better than the matching filter approach.
- 4 February 2022
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 87 (3) , U109-U120
Wave-equation migration velocity analysis via the optimal-transport-based objective function
Wave-equation migration velocity analysis builds a kinematically accurate macrovelocity model containing the large-scale structures of the model for seismic imaging or full-waveform inversion (FWI). Differential semblance optimization formulates the misfit function in the subsurface domain by applying a penalty to the unfocused subsurface-offset gathers or the nonflattened angle gathers. Such a penalty leads to gradients with strong artifacts and spurious oscillations, which then leads to slow convergence. We formulate the misfit function using the optimal transport (OT) among neighboring traces in angle-domain common-image gathers (CIGs). Specifically, we measure the unflatness of the gathers by comparing the Wasserstein distance of adjacent traces. Because CIGs are ideal to form a probability distribution, which is required by OT, it is natural to use OT to measure the time shifts between two distributions. Our objective function exhibits well-behaved convex and unimodal properties in a simple model with a single interface, and it is expected to be minimum for the correct velocity when the angle gathers are flat. A synthetic data set example on the Marmousi model indicates the ability of our method to provide an initial model for FWI that allows FWI to converge. We also apply the approach on a marine field data set to further demonstrate its effectiveness in estimating the macrovelocity model.
- 4 February 2022
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 87 (3) , U109-U120
Wave-equation migration velocity analysis via the optimal-transport-based objective function
Wave-equation migration velocity analysis builds a kinematically accurate macrovelocity model containing the large-scale structures of the model for seismic imaging or full-waveform inversion (FWI). Differential semblance optimization formulates the misfit function in the subsurface domain by applying a penalty to the unfocused subsurface-offset gathers or the nonflattened angle gathers. Such a penalty leads to gradients with strong artifacts and spurious oscillations, which then leads to slow convergence. We formulate the misfit function using the optimal transport (OT) among neighboring traces in angle-domain common-image gathers (CIGs). Specifically, we measure the unflatness of the gathers by comparing the Wasserstein distance of adjacent traces. Because CIGs are ideal to form a probability distribution, which is required by OT, it is natural to use OT to measure the time shifts between two distributions. Our objective function exhibits well-behaved convex and unimodal properties in a simple model with a single interface, and it is expected to be minimum for the correct velocity when the angle gathers are flat. A synthetic data set example on the Marmousi model indicates the ability of our method to provide an initial model for FWI that allows FWI to converge. We also apply the approach on a marine field data set to further demonstrate its effectiveness in estimating the macrovelocity model.
- 17 August 2021
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 86 (6) , T469-T485
Pseudoelastic pure P-mode wave equation
We have developed a pseudoelastic wave equation describing pure pressure waves propagating in elastic media. The pure pressure-mode (P-mode) wave equation uses all of the elastic parameters (such as density and the P- and S-wave velocities). It produces the same amplitude variation with offset (AVO) effects as PP-reflections computed by the conventional elastic wave equation. Because the new wave equation is free of S-waves, it does not require finer grids for simulation. This leads to a significant computational speedup when the ratio of pressure to S-wave velocities is large. We test the performance of our method on a simple synthetic model with high-velocity contrasts. The amplitude admitted by the pseudoelastic pure P-mode wave equation is highly consistent with that associated with the conventional elastic wave equation over a large range of incidence angles. We further verify our method’s robustness and accuracy using a more complex and realistic 2D salt model from the SEG Advanced Modeling Program. The ideal AVO behavior and computational advantage make our wave equation a good candidate as a forward simulation engine for performing elastic full-waveform inversion, especially for marine streamer data sets.
- 17 August 2021
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 86 (6) , T469-T485
Pseudoelastic pure P-mode wave equation
We have developed a pseudoelastic wave equation describing pure pressure waves propagating in elastic media. The pure pressure-mode (P-mode) wave equation uses all of the elastic parameters (such as density and the P- and S-wave velocities). It produces the same amplitude variation with offset (AVO) effects as PP-reflections computed by the conventional elastic wave equation. Because the new wave equation is free of S-waves, it does not require finer grids for simulation. This leads to a significant computational speedup when the ratio of pressure to S-wave velocities is large. We test the performance of our method on a simple synthetic model with high-velocity contrasts. The amplitude admitted by the pseudoelastic pure P-mode wave equation is highly consistent with that associated with the conventional elastic wave equation over a large range of incidence angles. We further verify our method’s robustness and accuracy using a more complex and realistic 2D salt model from the SEG Advanced Modeling Program. The ideal AVO behavior and computational advantage make our wave equation a good candidate as a forward simulation engine for performing elastic full-waveform inversion, especially for marine streamer data sets.
- 21 October 2020
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 85 (6) , R477-R492
ML-descent: An optimization algorithm for full-waveform inversion using machine learning
Full-waveform inversion (FWI) is a nonlinear optimization problem, and a typical optimization algorithm such as the nonlinear conjugate gradient or limited-memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) would iteratively update the model mainly along the gradient-descent direction of the misfit function or a slight modification of it. Based on the concept of meta-learning, rather than using a hand-designed optimization algorithm, we have trained the machine (represented by a neural network) to learn an optimization algorithm, entitled the “ML-descent,” and apply it in FWI. Using a recurrent neural network (RNN), we use the gradient of the misfit function as the input, and the hidden states in the RNN incorporate the history information of the gradient similar to an LBFGS algorithm. However, unlike the fixed form of the LBFGS algorithm, the machine-learning (ML) version evolves in response to the gradient. The loss function for training is formulated as a weighted summation of the L2 norm of the data residuals in the original inverse problem. As with any well-defined nonlinear inverse problem, the optimization can be locally approximated by a linear convex problem; thus, to accelerate the training, we train the neural network by minimizing randomly generated quadratic functions instead of performing time-consuming FWIs. To further improve the accuracy and robustness, we use a variational autoencoder that projects and represents the model in latent space. We use the Marmousi and the overthrust examples to demonstrate that the ML-descent method shows faster convergence and outperforms conventional optimization algorithms. The energy in the deeper part of the models can be recovered by the ML-descent even when the pseudoinverse of the Hessian is not incorporated in the FWI update.
- 21 October 2020
- Published by Society of Exploration Geophysicists in Geophysics
- Vol. 85 (6) , R477-R492
ML-descent: An optimization algorithm for full-waveform inversion using machine learning
Full-waveform inversion (FWI) is a nonlinear optimization problem, and a typical optimization algorithm such as the nonlinear conjugate gradient or limited-memory Broyden-Fletcher-Goldfarb-Shanno (LBFGS) would iteratively update the model mainly along the gradient-descent direction of the misfit function or a slight modification of it. Based on the concept of meta-learning, rather than using a hand-designed optimization algorithm, we have trained the machine (represented by a neural network) to learn an optimization algorithm, entitled the “ML-descent,” and apply it in FWI. Using a recurrent neural network (RNN), we use the gradient of the misfit function as the input, and the hidden states in the RNN incorporate the history information of the gradient similar to an LBFGS algorithm. However, unlike the fixed form of the LBFGS algorithm, the machine-learning (ML) version evolves in response to the gradient. The loss function for training is formulated as a weighted summation of the L2 norm of the data residuals in the original inverse problem. As with any well-defined nonlinear inverse problem, the optimization can be locally approximated by a linear convex problem; thus, to accelerate the training, we train the neural network by minimizing randomly generated quadratic functions instead of performing time-consuming FWIs. To further improve the accuracy and robustness, we use a variational autoencoder that projects and represents the model in latent space. We use the Marmousi and the overthrust examples to demonstrate that the ML-descent method shows faster convergence and outperforms conventional optimization algorithms. The energy in the deeper part of the models can be recovered by the ML-descent even when the pseudoinverse of the Hessian is not incorporated in the FWI update.
- 1 May 2020
- Published by Wiley in Geophysical Prospecting
- Vol. 68 (4) , 1154-1166
The effectiveness of a pseudo-inverse extended Born operator to handle lateral heterogeneity for imaging and velocity analysis applications
- 1 May 2020
- Published by Wiley in Geophysical Prospecting
- Vol. 68 (4) , 1154-1166
The effectiveness of a pseudo-inverse extended Born operator to handle lateral heterogeneity for imaging and velocity analysis applications
- Open Access
ML-misfit: Learning a robust misfit function for full-waveform inversion using machine learning
Most of the available advanced misfit functions for full waveform inversion (FWI) are hand-crafted, and the performance of those misfit functions is data-dependent. Thus, we propose to learn a misfit function for FWI, entitled ML-misfit, based on machine learning. Inspired by the optimal transport of the matching filter misfit, we design a neural network (NN) architecture for the misfit function in a form similar to comparing the mean and variance for two distributions. To guarantee the resulting learned misfit is a metric, we accommodate the symmetry of the misfit with respect to its input and a Hinge loss regularization term in a meta-loss function to satisfy the "triangle inequality" rule. In the framework of meta-learning, we train the network by running FWI to invert for randomly generated velocity models and update the parameters of the NN by minimizing the meta-loss, which is defined as accumulated difference between the true and inverted models. We first illustrate the basic principle of the ML-misfit for learning a convex misfit function for travel-time shifted signals. Further, we train the NN on 2D horizontally layered models, and we demonstrate the effectiveness and robustness of the learned ML-misfit by applying it to the well-known Marmousi model.
Open AccessML-misfit: Learning a robust misfit function for full-waveform inversion using machine learning
Most of the available advanced misfit functions for full waveform inversion (FWI) are hand-crafted, and the performance of those misfit functions is data-dependent. Thus, we propose to learn a misfit function for FWI, entitled ML-misfit, based on machine learning. Inspired by the optimal transport of the matching filter misfit, we design a neural network (NN) architecture for the misfit function in a form similar to comparing the mean and variance for two distributions. To guarantee the resulting learned misfit is a metric, we accommodate the symmetry of the misfit with respect to its input and a Hinge loss regularization term in a meta-loss function to satisfy the "triangle inequality" rule. In the framework of meta-learning, we train the network by running FWI to invert for randomly generated velocity models and update the parameters of the NN by minimizing the meta-loss, which is defined as accumulated difference between the true and inverted models. We first illustrate the basic principle of the ML-misfit for learning a convex misfit function for travel-time shifted signals. Further, we train the NN on 2D horizontally layered models, and we demonstrate the effectiveness and robustness of the learned ML-misfit by applying it to the well-known Marmousi model.
- 11 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 58 (7) , 4704-4720
Joint Minimization of the Mean and Information Entropy of the Matching Filter Distribution for a Robust Misfit Function in Full-Waveform Inversion
- 11 February 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 58 (7) , 4704-4720
Joint Minimization of the Mean and Information Entropy of the Matching Filter Distribution for a Robust Misfit Function in Full-Waveform Inversion
- Open Access
A data-driven choice of misfit function for FWI using reinforcement learning
In the workflow of Full-Waveform Inversion (FWI), we often tune the parameters of the inversion to help us avoid cycle skipping and obtain high resolution models. For example, typically start by using objective functions that avoid cycle skipping, like tomographic and image based or using only low frequency, and then later, we utilize the least squares misfit to admit high resolution information. We also may perform an isotropic (acoustic) inversion to first update the velocity model and then switch to multi-parameter anisotropic (elastic) inversions to fully recover the complex physics. Such hierarchical approaches are common in FWI, and they often depend on our manual intervention based on many factors, and of course, results depend on experience. However, with the large data size often involved in the inversion and the complexity of the process, making optimal choices is difficult even for an experienced practitioner. Thus, as an example, and within the framework of reinforcement learning, we utilize a deep-Q network (DQN) to learn an optimal policy to determine the proper timing to switch between different misfit functions. Specifically, we train the state-action value function (Q) to predict when to use the conventional L2-norm misfit function or the more advanced optimal-transport matching-filter (OTMF) misfit to mitigate the cycle-skipping and obtain high resolution, as well as improve convergence. We use a simple while demonstrative shifted-signal inversion examples to demonstrate the basic principles of the proposed method.
Open AccessA data-driven choice of misfit function for FWI using reinforcement learning
In the workflow of Full-Waveform Inversion (FWI), we often tune the parameters of the inversion to help us avoid cycle skipping and obtain high resolution models. For example, typically start by using objective functions that avoid cycle skipping, like tomographic and image based or using only low frequency, and then later, we utilize the least squares misfit to admit high resolution information. We also may perform an isotropic (acoustic) inversion to first update the velocity model and then switch to multi-parameter anisotropic (elastic) inversions to fully recover the complex physics. Such hierarchical approaches are common in FWI, and they often depend on our manual intervention based on many factors, and of course, results depend on experience. However, with the large data size often involved in the inversion and the complexity of the process, making optimal choices is difficult even for an experienced practitioner. Thus, as an example, and within the framework of reinforcement learning, we utilize a deep-Q network (DQN) to learn an optimal policy to determine the proper timing to switch between different misfit functions. Specifically, we train the state-action value function (Q) to predict when to use the conventional L2-norm misfit function or the more advanced optimal-transport matching-filter (OTMF) misfit to mitigate the cycle-skipping and obtain high resolution, as well as improve convergence. We use a simple while demonstrative shifted-signal inversion examples to demonstrate the basic principles of the proposed method.