Feb 13 2025 24 mins
“The biggest challenge for geophysicists? Learning machine learning's ‘new language’ from the world of statistics.”
Machine learning is transforming geoscience, and Gerard Schuster explains how. This conversation explores key ML applications in seismic interpretation, the role of convolutional neural networks in fault detection, and why hands-on labs are essential for mastering these techniques. With real-world examples and insights from his new book, Machine Learning Methods in Geoscience, this episode delivers practical knowledge for integrating ML into geophysics.
KEY TAKEAWAYS
> Why ML matters for geoscientists – The demand for ML skills is growing, and Jerry shares how this shift shapes education and careers.
> CNNs in action – Convolutional neural networks are used to detect rock cracks in Saudi Arabia through drone imagery.
> Transformers vs. traditional neural networks – Transformers process seismic data differently by capturing long-range dependencies, offering new advantages.
NEXT STEP
Explore Machine Learning Methods in Geoscience by Gerard Schuster, featuring hands-on MATLAB and Colab labs. Get the book and start applying ML techniques today! https://library.seg.org/doi/epdf/10.1190/1.9781560804048.fm
TEXT A FRIEND
These are great insights on how ML is actually being used in seismic work, not just theory. https://seg.org/podcasts/episode-249-machine-learning-methods-in-geoscience
GUEST BIO
Gerard Schuster has an M.S. (1982) and a Ph.D. (1984) from Columbia University and was a postdoctoral researcher there from 1984 to 1985. From 1985 to 2009, he was a professor of geophysics at the University of Utah and became a professor of geophysics at KAUST (2009–2021). He is currently a research professor at the University of Utah. He received several teaching and research awards while at the University of Utah. He was editor of GEOPHYSICS 2004–2005 and was awarded SEG’s Virgil Kauffman Gold Medal in 2010 for his work in seismic interferometry. His previous books are Seismic Interferometry (2009, Cambridge Press) and Seismic Inversion (2017, SEG).
LINKS
* Buy the Print Book at https://seg.org/shop/product/?id=fe5a3cd3-77b2-ef11-b8e8-6045bda82e05
* Visit https://seg.org/podcasts/episode-249-machine-learning-methods-in-geoscience for the full guest bios and show notes.
CALL FOR ABSTRACTS
Technical Program Chairs Yingcai Zheng and Molly Turko invite you to submit your best work. This year, we're fostering deeper collaboration between SEG, AAPG, and SEPM. Focus on regional challenges and how integrated geoscience can unlock solutions.
Submit short or expanded abstracts for oral and poster presentations. The Call for Abstracts is open and closes on 15 March at 5:00 PM CT.
Don't miss this opportunity to share your research and connect with the broader geoscience community at https://www.imageevent.org/.
SHOW CREDITS
Andrew Geary at TreasureMint hosted, edited, and produced this episode. The SEG podcast team comprises Jennifer Cobb, Kathy Gamble, and Ally McGinnis.
If you have episode ideas or feedback for the show or want to sponsor a future episode, email the show at [email protected].
Machine learning is transforming geoscience, and Gerard Schuster explains how. This conversation explores key ML applications in seismic interpretation, the role of convolutional neural networks in fault detection, and why hands-on labs are essential for mastering these techniques. With real-world examples and insights from his new book, Machine Learning Methods in Geoscience, this episode delivers practical knowledge for integrating ML into geophysics.
KEY TAKEAWAYS
> Why ML matters for geoscientists – The demand for ML skills is growing, and Jerry shares how this shift shapes education and careers.
> CNNs in action – Convolutional neural networks are used to detect rock cracks in Saudi Arabia through drone imagery.
> Transformers vs. traditional neural networks – Transformers process seismic data differently by capturing long-range dependencies, offering new advantages.
NEXT STEP
Explore Machine Learning Methods in Geoscience by Gerard Schuster, featuring hands-on MATLAB and Colab labs. Get the book and start applying ML techniques today! https://library.seg.org/doi/epdf/10.1190/1.9781560804048.fm
TEXT A FRIEND
These are great insights on how ML is actually being used in seismic work, not just theory. https://seg.org/podcasts/episode-249-machine-learning-methods-in-geoscience
GUEST BIO
Gerard Schuster has an M.S. (1982) and a Ph.D. (1984) from Columbia University and was a postdoctoral researcher there from 1984 to 1985. From 1985 to 2009, he was a professor of geophysics at the University of Utah and became a professor of geophysics at KAUST (2009–2021). He is currently a research professor at the University of Utah. He received several teaching and research awards while at the University of Utah. He was editor of GEOPHYSICS 2004–2005 and was awarded SEG’s Virgil Kauffman Gold Medal in 2010 for his work in seismic interferometry. His previous books are Seismic Interferometry (2009, Cambridge Press) and Seismic Inversion (2017, SEG).
LINKS
* Buy the Print Book at https://seg.org/shop/product/?id=fe5a3cd3-77b2-ef11-b8e8-6045bda82e05
* Visit https://seg.org/podcasts/episode-249-machine-learning-methods-in-geoscience for the full guest bios and show notes.
CALL FOR ABSTRACTS
Technical Program Chairs Yingcai Zheng and Molly Turko invite you to submit your best work. This year, we're fostering deeper collaboration between SEG, AAPG, and SEPM. Focus on regional challenges and how integrated geoscience can unlock solutions.
Submit short or expanded abstracts for oral and poster presentations. The Call for Abstracts is open and closes on 15 March at 5:00 PM CT.
Don't miss this opportunity to share your research and connect with the broader geoscience community at https://www.imageevent.org/.
SHOW CREDITS
Andrew Geary at TreasureMint hosted, edited, and produced this episode. The SEG podcast team comprises Jennifer Cobb, Kathy Gamble, and Ally McGinnis.
If you have episode ideas or feedback for the show or want to sponsor a future episode, email the show at [email protected].