Pau Batlle
PhD Candidate, California Institute of Technology
I am a PhD candidate in Computing and Mathematical Sciences at Caltech, where I am fortunate to be advised by Houman Owhadi.
Prior to joining Caltech, I graduated from Universitat Politècnica de Catalunya with a double degree in Mathematics and Engineering Physics and I was a research intern at the Center for Data Science at NYU. I have done research at different institutions and companies, which can be found in my résumé
My current research areas include Statistics, Uncertainty Quantification and Gaussian Processes, both from a theoretical point of view and applications. I have worked in different domain applications including remote sensing, biology, earthquake prediction, epidemic modelling and telecommunications engineering. I published in journals and conferences across physics, diverse areas of applied mathematics and machine learning, see my publications.
Recent news
I gave a talk about the disproof of the Burrus conjecture at SIAM UQ 2024. Slides are available here - March 2024
New preprint about learning hypergraph function structure from data using Gaussian Processes! - December 2023
My paper Kernel Methods are competitive for Operator Learning has been published in the Journal of Computational Physics (JCP) - November 2023
I passed my candidacy! Find the slides from my candidacy talk here - October 2023here
My new preprint about constrained inference in inverse problems, including a disproof of the Burrus conjecture, is now out! Optimization-based frequentist confidence intervals for functionals in constrained inverse problems: Resolving the Burrus conjecture. Slides and a poster are available - October 2023
I gave a talk about my paper on kernel methods for operator learning at ICIAM 2023, in the "Machine Learning in infinite dimensions" minisymposium. Slides are available here - August 2023