Machine Learning (ML) is a subfield of Artificial Intelligence (AI) that has shown recently incredible potential in solving problems efficiently and providing a measure of certainty to the solution. It has experienced rapid growth in the last decade across diverse industries, including communications, financial services, security, transportation, and others. A combination of large-scale data, a new array of Machine Learning tools, robust mathematical foundations, big advances in computer hardware and software led to the development of Machine Learning applications which produced dramatic results, enabling new opportunities. Among these opportunities are finding solutions to the many Earth discovery problems. From seismic to well, gravity and magnetics, and all other measurements of the Earth physical properties, the geosciences community is now realizing the potential of machine learning in building a realistic model of the Earth from Oil & Gas exploration and production, to geotechnical and environmental applications.

This first of its kind of workshop in KAUST (combining Machine Learning with Earth sciences) aims to highlight the challenges, opportunities, and trends related to the adoption of Machine Learning in geosciences research and industrial workflows. Researchers from academia, geological bodies, Oil & Gas, and technology companies will present applications and case studies, promote discussion, and propose practical solutions to take greater advantage of Machine Learning methods.

The above are closely related to the "energy" and "digital" pillars of the KAUST research thrusts. The combined effort in pushing the boundaries on both domains is widely acknowledged in both the research community and the industry.

 

Topics that will be covered

  • State of the Art Approaches for Geophysical ML Applications
  • ML/AI-powered subsurface inversion/interpretation - Case studies
  • Using Big Data to Reduce Seismic Imaging Uncertainty
  • Geophysical data preparation for ML
  • Subsurface Analytics - Harnessing massive volumes of data
  • Using ML to optimize workflows
  • Global vs. reservoir cell scale ML
  • New approaches to seismic interpretation using machine learning
  • Geophysical methods in ML framework
  • Full-waveform inversion and ML
  • Deep learning vs. shallow networks (aka seismic inversion)

 

Organizing Committee

Tariq Alkhalifah [King Abdullah University of Science and Technology]

Xiangliang Zhang [King Abdullah University of Science and Technology]

Daniel Peter [King Abdullah University of Science and Technology]

Hussein Hoteit [King Abdullah University of Science and Technology]

Saber Feki  [King Abdullah University of Science and Technology]

 

Organizational Support

Christos Tzivanakis

 

For any queries contact us at swag@kaust.edu.sa