Résumé / Abstract Journal-club_Doctorants

Séminaire Doctorants / Seminar PhD students

« Simulating galaxy survey images to infer the properties of galaxies through cosmic times »

Sébastien Carassou
Institut d'Astrophysique de Paris (Paris, France)

After 20 years of large scale galaxy surveys (SDSS, CFHTLS, UDF etc.), the field of galaxy evolution has entered the realm of Big Data. We can now extract the spectrophotometric and morphometric properties of millions of galaxies, over a period that covers more than 9 billion years of cosmic history. But current constrains on models of galaxy evolution suffer from selection biases that, if not taken into account carefully, can lead to contradictory predictions.
To adress this issue, we (I!) are developing a new approach combining machine learning techniques (using a Bayesian framework) and empirical modeling with realistic image simulation using measured point-spread functions that reproduce a large fraction of these selection effects. This will allow us to perform a direct comparison between observed and simulated images and therefore to infer robust constraints on model parameters predicting the evolution of bulges and disks from z~2 to z~0.
mercredi 25 janvier 2017 - 17:00
Salle Entresol Daniel Chalonge, Institut d'Astrophysique
Page web du Séminaire / Seminar's webpage