Miguel Corrales

Ph.D. Students

Ph.D. Student - Hoteit Group

Research Interests

Miguel’s research is focused on Reservoir characterization, Carbon Capture and Storage, and Deep Learning.

Selected Publications

  • Corrales, M., Izzatullah, M., Ravasi, M., & Hoteit, H. (2022, August). Bayesian RockAVO: Direct petrophysical inversion with hierarchical conditional GANs. In Second International Meeting for Applied Geoscience & Energy (pp. 2253-2257). Society of Exploration Geophysicists and American Association of Petroleum Geologists.
  • Corrales, M., Ravasi, M., & Hoteit, H. (2022, June). Data-Driven, Direct Rock-Physics Inversion of Pre-Stack Seismic Data. In 83rd EAGE Annual Conference & Exhibition (Vol. 2022, No. 1, pp. 1-5). European Association of Geoscientists & Engineers.
  • Corrales, M., Mantilla Salas, S., Tasianas, A., Hoteit, H., & Afifi, A. (2022, February). The Potential for Underground CO2 Disposal Near Riyadh. In International Petroleum Technology Conference. OnePetro.


  • M.Sc., Energy Resources and Petroleum Engineering, KAUST, Thuwal, Saudi Arabia, 2019 – 2021
  • B.Sc., Petroleum Engineering, Universidad Central del Ecuador (UCE), Quito, Ecuador, 2012 - 2017

Professional Profile

  • 2019 - Current: MS Student, King Abdullah University of Science and Technology (KAUST), Thuwal, Saudi Arabia.
  • 2017: Intern, Petroamazonas EP, Quito, Ecuador.
  • 2016: Intern, Unidad de Servicios para la Industria Petrolera (USIP - UNAM), Mexico City, Mexico.
  • 2016: Exchange Student, Universidad Nacional Autonoma de Mexico (UNAM), Mexico City, Mexico.

Scientific and Professional Membership

  • Society of Petroleum Engineers (SEP)
  • European Association of Geoscientists and Engineers (EAGE)
  • Society of Exploration Geophysicists (SEG)


  • ​UCE Scholarship, Academic Excellence. 2012-2016

KAUST Affiliations

Ali- Al Naimi Petroleum Engineering Research Center (ANPERC)

Division of Physical Science and Engineering (PSE)

Deep Imaging Group (DIG)

Research Interests Keywords

Reservoir Characterization CCS Deep learning Generative modeling