Hi! I’m Pablo, physicist and deep learning scientist. In 2021 I completed my PhD in Physics at the University of València and Instituto de Física Corpuscular, investigating the nature of dark matter through cosmology, specially via the 21 cm cosmological signal of the hydrogen. In my research, I have explored different dark matter candidates, such as primordial black holes, devoting several works to examine the bounds on their abundance based on 21 cm cosmology and neutrino fluxes.

I have applied deep learning techniques to cosmology, such as convolutional neural networks and graph neural networks. For instance, in this work I have trained a model to weigh the masses of our galaxy and its neighbor Andromeda, being the first time that artificial intelligence is applied to this task. I’m also member of the CAMELS collaboration, a project to produce cosmological simulations specially designed for machine learning applications.

Since 2022 I am working at the Computer Vision Center at the Universitat Autònoma de Barcelona as a deep learning scientist, within the CARLA project, an autonomous driving simulator. There, I have developed different machine learning and computer vision tasks, including accelerating terramechanics simulations via convolutional and graph neural networks, training models for semantic segmentation, validating the implementation of wide lens cameras and implementing driving AI agents.

I have led several projects with international collaborators, written 20 scientific articles and given talks at the universities of Princeton, Tokyo, Nagoya and Brussels. You can see below one of my recent talks on deep learning applications in cosmology.

For more details, see here my Curriculum Vitae.