Jan. 23, 2019


Description

 
 
Generative adversarial networks in seismic data processing (Stephen Alwon, Schlumberger)

Generative adversarial networks (GANs) are a class of machine learning techniques that involve two networks trained simultaneously to generate a desired outcome. These schemes have had success in many traditional image processing tasks, such as style transfer and super-resolution, but are relatively unexplored in geophysics. We outline the underlying theory behind GANs and present networks that can perform traditional seismic processing tasks such as noise attenuation and trace interpolation.

Convolutional neural network for Salt model building in the Gulf of Mexico (Sribharath Kainkaryam, TGS )

Salt model building is one of the most time consuming, labor intensive and difficult to automate processes in the entire depth imaging workflow requiring significant human effort. The challenge and need for automating salt interpretation is well recognized by the seismic imaging community and has seen many applications of deep learning based convolutional neural network (CNN) architectures to carry out this task. However, significant challenges remain for reliable production-scale deployment of CNN based methods for salt model building owing to poor generalization capabilities of these networks. When used on new surveys, never seen by the CNN models during the training stage, the interpretation accuracy of these models drops significantly. To remediate this key problem, we introduce a U-shaped encoder-decoder type CNN architecture and use a specialized regularization strategy aimed at reducing the generalization error of the network. Our regularization scheme perturbs the ground truth labels in the training set. Two different perturbations are discussed: one which randomly changes the labels of the training set, flipping salt labels to sediments and vice versa and the second which smooths the labels.  We demonstrate that such perturbations act as a strong regularizer preventing the network from making low entropy, highly confident predictions on the training set and thus reducing overfitting. An ensemble strategy is also devised for test time augmentation that is shown to further improve the accuracy. The robustness, in terms of vastly improved generalization capability as well as improved interpretation accuracy of our training strategy is demonstrated with real data examples from the Gulf of Mexico. 

 

Sponsors:

 

 


Featured Speakers

Speaker: Stephen Alwon
Speaker Stephen Alwon
Stephen Alwon is a Geophysicist at Schlumberger with over 10 years experience in seismic imaging. He has been involved with all aspects of seismic data work, from acquisition and processing to model building and imaging. Although his primary work focuses on model building and imaging, he has recently begun exploring ...

Stephen Alwon is a Geophysicist at Schlumberger with over 10 years experience in seismic imaging. He has been involved with all aspects of seismic data work, from acquisition and processing to model building and imaging. Although his primary work focuses on model building and imaging, he has recently begun exploring various ways machine learning can impact the geosciences.


 

Full Description

Speaker: Sribharath Kainkaryam
Speaker Sribharath Kainkaryam

Sribharath (Sri) Kainkaryam is a data scientist with TGS. He works on applications of deep learning and machine learning problems in seismic imaging and petrophysics. He has a background in seismic imaging research having worked as a research scientist for Schlumberger working on seismic imaging and velocity model building. 


View all speakers

Organizer

Data Analytics Study Group

Andrés Landa


Date and Time

Wed, Jan. 23, 2019

6 p.m. - 7:30 p.m.
(GMT-0500) US/Central

Event has ended

If you do not have a full-time job in the oil and gas industry, are a full-time student or Member in Transition (MiT) member, and you do not see a discounted registration fee for students/MiT regarding this event, please contact the GCS manager at spe-gcs@spe.org.


Location

Intelie - Rignet Office

15115 Park Row Dr. Ste 300
Houston, Texas 77084
United States



Group(s): Data Analytics