No dia 12 de março, o Departamento de Física do ISEL promove o seminário "Numerical Earth models".

O evento, que decorrerá pelas 14h00 no Auditório A, terá como orador Leonardo Azevedo (Instituto Superior Técnico – Universidade de Lisboa).
 

Abstract:

Numerical Earth models are central pieces of information for well-informed decision-making for sustainable exploration and production of mineral and energy resources.

These models should be able to reproduce expected geological settings and assimilate multiple data with various origin, quality and resolution. This is a rather challenging task that often involves solving a geophysical inverse problem.

Deep generative learning methods comprise a set of flexible tools that allow the simultaneous integration of data while predicting geologically plausible earth models.

This talk review the main concepts of spatial data science and deep generative learning with several applications in real case studies.