BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:pop-cosmos: Comprehensive Forward Modelling of Photometric Galaxy 
 Survey Data - Stephen Thorp - Oskar Klein Centre\, Stockholm University
DTSTART:20240327T100000Z
DTEND:20240327T110000Z
UID:TALK213505@talks.cam.ac.uk
CONTACT:Sinan Deger
DESCRIPTION:Projects such as the imminent Vera C. Rubin Observatory are cr
 itical tools for understanding cosmological questions like the nature of d
 ark energy. By observing huge numbers of galaxies\, they enable us to map 
 the large scale structure of the Universe. To do this\, however\, we need 
 reliable ways of estimating galaxy redshifts from only photometry. I will 
 present an overview of our pop-cosmos forward modelling framework for phot
 ometric galaxy survey data\, a novel approach which connects photometric r
 edshift inference to a physical picture of galaxy evolution. Within pop-co
 smos\, we model galaxies as draws from a population prior distribution ove
 r redshift\, mass\, dust properties\, metallicity\, and star formation his
 tory. These properties are mapped to photometry using an emulator for stel
 lar population synthesis (speculator/photulator)\, followed by the applica
 tion of a learned model for a survey's noise properties. Application of se
 lection cuts enables the generation of mock galaxy catalogues. This natura
 lly enables us to use simulation-based inference to solve the inverse prob
 lem of calibrating the population-level prior on physical parameters from 
 a deep photometric galaxy survey. The resulting model can then be used to 
 derive accurate redshift distributions for upcoming photometric surveys\, 
 for instance for facilitating weak lensing and clustering science. We use 
 a diffusion model as a flexible population-level prior\, and optimise its 
 parameters by minimising the Wasserstein distance between forward-simulate
 d photometry and the real survey data. I will show applications of this fr
 amework to COSMOS data\, and will demonstrate how we are able to extract t
 he redshift distribution\, and make inference about galaxy physics\, from 
 our learned population prior.
LOCATION:Martin Ryle Seminar Room\, KICC
END:VEVENT
END:VCALENDAR