Diego Villalba

Physics · Machine Learning · Scientific Computing

I am a physicist working at the intersection of machine learning and scientific computing. My research focuses on cosmology, computational astrophysics, and the data-driven modeling of physical systems — with the goal of developing AI methods that are scalable, interpretable, and physically grounded.

I am particularly drawn to the idea that machine learning and physics are not separate disciplines, but complementary frameworks that can be unified to address complex scientific, technological, and societal challenges. This perspective guides both my research and teaching.

Cosmological ML

Displacement-field reconstruction and super-resolution of N-body simulations using neural implicit representations.

Medical Imaging

Accelerating MRI acquisition with wavelet scattering transforms while preserving diagnostic fidelity.

Differentiable Models

Simulation-based inference and neural fields for high-dimensional physical field reconstruction.

Teaching

Co-teaching data mining at the Facultad de Ciencias, UNAM — connecting theory with applied ML.

I have developed these methods at the Donostia International Physics Center (DIPC), where I contributed to the CODECS project on cosmological simulation compression, and at the Universidad Nacional Autónoma de México (UNAM), where my ongoing work spans astrophysical super-resolution and interpretable MRI reconstruction.

Applied domains

Medical Imaging

Accelerating MRI acquisition while maintaining diagnostic image quality.

Industry

OCR-based data processing pipelines and document intelligence systems.

Research Interests

Machine learning for scientific simulations

Simulation-based inference

Cosmological simulations & large-scale structure

Differentiable and physics-informed modeling

Machine learning for medical imaging

High-performance & parallel computing

Neural implicit representations

Generalizable methods for real-world systems

Personal

Originally from Mexico City. Alongside research, I am actively involved in teaching, mentoring, and science communication, and I enjoy helping students build strong foundations in physics and data science.

I am always interested in connecting with people working on machine learning, scientific computing, or data-driven science — both in academia and industry.

Explore my research work, projects, and CV — or reach out if you'd like to collaborate.