2026 workshop program
Data-Driven High-Dimensional Inverse Problems: A Journey Through Strong Gravitational Lensing Data Analysis
The work
Abstract
While upcoming observatories promise to enable percent-level precision science across many areas of astrophysics, realizing this potential hinges on our ability to develop new analysis methods that can meet the scale and complexity of the data. In many cases, traditional analysis methods can fall short of fully extracting the information contained in observations. Strong gravitational lensing, a nonlinear, non-convex, and high-dimensional inverse problem, exemplifies these challenges and opportunities. This talk explores recent advances in Bayesian image reconstruction and high-dimensional inference, with a focus on applications to lensing. Using score-based generative approaches, I will present methods for reconstructing high-fidelity astronomical images while addressing key challenges such as out-of-distribution data robustness, uncertainty quantification, and joint inference of hierarchical properties. I will also introduce statistical tools to assess the accuracy of posterior samples obtained with machine learning. These advances in deep learning-based inference open new possibilities for studying complex systems at scale, particularly in the era of large surveys like LSST, Euclid, and the Roman Space Telescope.
The speaker
Biography
Laurence Perreault-Levasseur is an associate professor at the University of Montréal and a Core Mila Member, where she conducts research in the development and application of machine learning methods to cosmology, with a special emphasis on strong gravitational lensing and other high-dimensional inverse problems. She is also a Visiting Scholar at the Flatiron Institute in New York City. Prior to that, she was a Flatiron research fellow at the Center for Computational Astrophysics in the Flatiron Institute and a KIPAC postdoctoral fellow at Stanford University. Laurence completed her PhD degree at the University of Cambridge, where she worked on applications of open effective field theory methods to the formalism of inflation. She received her B.Sc. and M.Sc. degrees from McGill University.