Séminaires du DPhP: Rubin/LSST : the instrument, the survey, and machine learning approaches to weak-lensing analyses.

par Eric AUBOURG (APC)

lundi 19 janv. 2026, 11:00 → 12:30 Europe/Paris

CEA Paris-Saclay

The Vera C. Rubin Observatory’s Legacy Survey of Space and Time (LSST) will deliver an unprecedented, deep and wide optical imaging dataset of the southern sky, combining a powerful 8.4-meter telescope, a wide-field camera, and a decade-long observing strategy optimized for time-domain discovery and precision cosmology. In this seminar I will give an overview of Rubin/LSST: the observatory and instrument design, the survey strategy and data products, and its broad science program.

I will then focus on one of LSST’s flagship cosmology probes: weak gravitational lensing, where percent-level control of systematics and robust inference from complex, crowded imaging are essential. I will highlight how machine-learning approaches can help mitigate source confusion (blending) in deep, high-density fields, thereby reducing biases in shear and redshift estimation. I will then discuss our vision for Bayesian end-to-end pipelines that propagate uncertainties from pixels to cosmological constraints, and the role of simulation-based inference.

 

zoom connection: https://cern.zoom.us/j/68399739111?pwd=SGErcHBQYmVLWFZTVE9jdm5xSmdEdz09