Pablo Lemos

Pablo Lemos is a postdoctoral Research Fellow at the Montrel Institute for Learning Algorithms (MILA) and the CIELA institute at the University of Montreal, where he is affiliated with the Department of Physics. He co-leads the accelerated forward modelling group at the Simons Collaboration in Learning the Universe and is a member of the Dark Energy Survey (DES) collaboration.

Research Focus

Lemos specializes in applying machine learning techniques to astrophysics and cosmology problems. His core research areas include:

• Graph neural networks for cosmological simulations
• Symbolic regression methods for data analysis
• Simulation-based inference techniques
• Bayesian inverse problem solving
• Large-scale structure analysis

Publication Analysis

Lemos's recent publications (2024-2025) demonstrate strong interdisciplinary work combining cosmology, astrophysics, and machine learning. Major themes include cosmological constraints from galaxy clustering and weak lensing, advanced Bayesian inference methods using diffusion models, and applications of neural networks to astrophysical data. Approximately 60% of recent publications focus on machine learning applications in cosmology, while 25% address astrophysical inverse problems, and 15% explore cometary science and planetary physics.

Collaborations and Groups

• Co-leader of accelerated forward modelling group at Simons Collaboration in Learning the Universe
• Active member of Dark Energy Survey (DES) collaboration