Cognite and National Instruments are teaming up this month to explain why machine learning on its own isn't a silver bullet for the oil and gas industry.
Machine learning has been immensely successful in areas such as image recognition, optimizing ad revenue, and recommending news feeds based on customer preferences and product positioning. But machine learning has not had the same impact in oil and gas, even though the value potential is extraordinarily high.
The reason why: The dynamics in the oil and gas industry are governed by the laws of physics, which are substantially different from typical consumer behavior.
In a hands-on workshop at Rice University on Oct. 14 from 10:00 a.m. to noon, attendees will learn why machine learning seems to fail in the oil and gas industry -- and how to fix it! Topics to be covered include how machine learning can be combined with physics-based models to produce reliable predictions, robustness, solid extrapolation properties, and thus real value.
Who is this workshop for?
The workshop will be hosted by Gunnar Staff, Senior Principal Data Scientist at Cognite.
Staff leads a team of 14 data scientists working to optimize the production at oil and gas installations around the world. Before joining Cognite, Gunnar spent 12 years as a Principle R&D Software Engineer at Schlumberger, where he worked on the core numerics and physics of the OLGA Dynamic Multiphase Flow Simulator. Gunnar has published numerous papers on multiphase flow modeling.
Interested in attending? RSVP by Oct. 12 to firstname.lastname@example.org.