Applications of Machine Learning: Physics and Data Driven Seismic Data Analysis

Mrinal Sen, University of Texas at Austin

Description

Much of the seismic data analysis has been carried out using methods based on the physics of wave propagation and some signal processing principles. These methods are generally termed ‘physics-based’ or ‘model- based’ approaches. The entire suite of seismic processing and inversion algorithms starting from stacking velocity analysis to full waveform inversion is developed on these principles. Recently, a flurry of data-driven models that are agnostic to the physics has been developed for application to some previously solved and unsolved problems. Physics based approached rely on fundamental physical principles resulting in some predictive equations. However, it may suffer from the limitations of being incomplete. Data driven approaches, on the other hand, are methods for understanding the mechanisms generally by identifying patterns in large volumes of data. Thus, it may appear that we are at cross-roads to decide on the applicability and usefulness of the two seemingly different approaches. Even a broader question is “do we need to choose between the two?”

Questions answered in this course

• How do we relate input data to the desired output?
• What are the appropriate questions to ask?
• What is the role of physics in seismic data analysis? A historical perspective.
• Are we missing something in our physics-based models? What are the limitations, if any?
• Can the signal processing based approaches address some of the limitations?
• What is a data-driven or a machine learning (ML) approach?
• How do ML methods work? Is ML just the old wine in a new bottle?
• What is the role of physics in ML formulation?
• Do ML methods offer any advantages over current physics based approaches?

Goals

Like any other industry, seismic industry is abuzz with the resurgence of artificial intelligence and machine learning. Is this just a fashionable thing to do? Can we expect to revolutionize the seismic data processing and interpretation steps? The goal of this course is to take a step-by- step approach to explain the physics, signal processing and ML based approaches in seismic data analysis. The focus is not on the theoretical principles but only the applicability and usefulness of these approaches. Examples from basic data conditioning and NMO to velocity estimation, QSI and automated interpretation, will be used to demonstrate the current status and future directions.

Who Should Attend

The course is intended to all practitioners including R&D professionals, managers, data processors and interpreters. The primary target audience is exploration geophysicists, who are interested in not just an overview of the new data driven technologies but the intellectual merits and value addition in our trade.

• Processing geophysicists would find this course useful to choose appropriate methods
• R&D professionals would be benefited from learning the details
• Interpreters would be able to appreciate the ease of large scale data exploration
• Managers would be able to assess the value addition of this new technology

Course Book

Course attendees receive the book as part of the registration fee. If you are unable to attend the DISC course but are interested in the book, it can be purchased separately in print or as an ebook.

Biography

Mrinal Sen

View Bio

Close

Mrinal Sen

Mrinal K. Sen is a professor of Geophysics and holder of Morgan Davis Centennial chair in Petroleum Geology at the Department of Earth and Planetary Sciences and the Institute for Geophysics at the University of Texas at Austin. During 2013 and 2014, Mrinal served as the director of the National Geophysical Research Institute, Hyderabad, India. He received his M.Sc degree from IIT Dhanbad and PhD from the University of Hawaii at Manoa, USA.

Mrinal is known internationally for his work on theoretical and computational seismology, and geophysical inversion He has authored a textbook that was nominated for an award, co-authored another widely used textbook that is now in its second edition at Cambridge University Press, taught multiple short courses including one on FWI for SEG, and authored or co-authored 20 book chapters and over 200 journal articles. Mrinal has received many awards, including the Honorary membership in the Society of Exploration Geophysicists “for extraordinary contributions as a geophysicist, educator and author,” the ‘Joseph C. Walter award for research excellence,’ the ‘distinguished educator award’ at the University of Texas, Decentennial Gold Medal of the Indian Geophysical Union, and the distinguished alumnus award from IIT(ISM) and the University of Hawaii at Manoa. He is the recipient of 2018 Virgil Kauffman gold medal of the Society of Exploration Geophysicists for making significant advancement in the sciences of exploration geophysics in the last five years.

Schedule

Check back soon for more tour stops!

If you are interested in hosting our 2024–2025 DISC, please contact us.