Online only event - You must pre-register to receive access information.

You must be logged in to register.

Speaker: Altay Sansal, Quantico Energy Solutions

In this talk, after a brief introduction to Machine Learning (ML), Altay Sansal takes us through a novel ML workflow using Neural Networks applied to Quantitative Seismic Interpretation (QI). Topics covered will include modern techniques like data augmentation, feature engineering, model parameter optimization, and combining the power of multiple models to improve generalization of the final predictive model. In sparse data environments like QI, these methods help us achieve high-resolution rock/fluid properties in a short period of time with minimal prior information and user intervention.

Speaker Biography: Altay Sansal, Quantico Energy Solutions
Altay Sansal has a master’s degree in Geophysics from University of Houston. He has years of experience in seismic imaging, quantitative interpretation (QI) and has extensive knowledge in modeling, statistics, machine learning, and inverse theory. Altay was a part of QI and seismic imaging teams at Geotrace Technologies, Geokinetics, and SAExploration. He managed QI teams at Geokinetics and SAExploration and was leading the R&D advisory boards in both companies. After this experience, he joined Quantico Energy Solutions and is currently working on applications of Neural Networks to quantitative interpretation. His academic work with Dr. John Castagna was titled “Analyzing Relevance of Direct Hydrocarbon Indicators in Prospect Risk Analysis” using data analytics techniques.