Luz Angélica Caudillo Mata, PhD

Bio

I am a Computational Scientist and Applied Mathematician with a demonstrated history of working as a researcher in Academia, Government Research Centers, and (more recently) Industry. I love being part of interdisciplinary teams that focus on solving problems whose solutions benefit our society. I am also a bridge builder that works to empower the communities I belong to by starting or organizing programs, mostly in the form of events.

Currently, I work as an Industrial Posdoctoral Fellow at the University of British Columbia (UBC), Vancouver campus, in Prof. Eldad Haber’s group. As part of my posdoc, I collaborate with the company Computational Geosciences, Inc as a member of their Artificial Intelligence reserch team. My project focuses on exploring convolutional neural network approaches for aquifer prospectivity mapping.

I hold a Ph.D. in Geophysics and Applied Mathematics from UBC, a M.Sc. in Computing and Industrial Mathematics from the Mathematics Research Center (CIMAT), and a B.Sc. in Computational Mathematics from the University of Guanajuato. Prior to earning my doctoral degree, I worked for 5+ years as a (C++) software developer at CIMAT. Throughout my studies, I conducted 5 internships at various research institutions in USA and Spain, including Lawrence Berkeley and Livermore National Laboratories.

Research interests

I conduct research in how to design and apply numerical methods in the following topics:

• convolutional neural networks for aquifer prospectivity mapping.
• upscaling and multiscale methods to simulate large-scale geophysical electromagnetic forward and inverse problems in 3D. Applications of this research occur in the detection and characterization of mineral deposits and petroleum reservoirs.
• mimetic multiscale methods for efficient simulation of electromagnetic fields in geophysical scenarios that include metallic-cased boreholes and fractures filled with conductive/resistive fluids. Applications of this research occur in the context of Hydraulic fracturing.

My most relevant areas of expertise are:

• Partial differential equations (PDE’s) (Maxwell’s equations)
• PDE constrained optimization
• Discretization of PDE’s (finite volume, finite element)
• Sparse linear algebra
• Numerical upscaling methods
• Mimetic Multiscale (finite element and finite volume) methods
• Geophysical electromagnetic forward and inverse problems
• Large-scale computation (high performance computing)
• (More recently) Convolutional Neural Networks