Title: Deblending galaxies with variational autoencoders: A joint multiband, multi-instrument approach

Abstract

ABSTRACT Blending of galaxies has a major contribution in the systematic error budget of weak-lensing studies, affecting photometric and shape measurements, particularly for ground-based, deep, photometric galaxy surveys, such as the Rubin Observatory Legacy Survey of Space and Time (LSST). Existing deblenders mostly rely on analytic modelling of galaxy profiles and suffer from the lack of flexible yet accurate models. We propose to use generative models based on deep neural networks, namely variational autoencoders (VAE), to learn probabilistic models directly from data. We train a VAE on images of centred, isolated galaxies, which we reuse, as a prior, in a second VAE-like neural network in charge of deblending galaxies. We train our networks on simulated images including six LSST bandpass filters and the visible and near-infrared bands of the Euclid satellite, as our method naturally generalizes to multiple bands and can incorporate data from multiple instruments. We obtain median reconstruction errors on ellipticities and r-band magnitude between ±0.01 and ±0.05, respectively, in most cases, and ellipticity multiplicative bias of 1.6 per cent for blended objects in the optimal configuration. We also study the impact of decentring and prove the method to be robust. This method only requires the approximate centre of eachmore » target galaxy, but no assumptions about the number of surrounding objects, pointing to an iterative detection/deblending procedure we leave for future work. Finally, we discuss future challenges about training on real data and obtain encouraging results when applying transfer learning.« less

Authors:

Arcelin, Bastien  ^[1];

• Search DOE PAGES for author "Arcelin, Bastien"
• Search DOE PAGES for ORCID "0000-0002-3553-792X"

Doux, Cyrille  ^[2];

• Search DOE PAGES for author "Doux, Cyrille"
• Search DOE PAGES for ORCID "0000-0003-4480-0096"

Aubourg, Eric ^[1]; Roucelle, Cécile ^[1]; The LSST Dark Energy Science Collaboration) ^[3]

---

+ Show Author Affiliations

1. Université de Paris, CNRS, Astroparticule et Cosmologie, F-75013 Paris, France
2. Université de Paris, CNRS, Astroparticule et Cosmologie, F-75013 Paris, France, Center for Particle Cosmology, Department of Physics and Astronomy, University of Pennsylvania, 209 South 33rd Street, Philadelphia, PA 19104, USA
3. (

Publication Date:

Thu Oct 08 00:00:00 EDT 2020

Sponsoring Org.:

USDOE

OSTI Identifier:

1734396

Grant/Contract Number:  

AC02-76SF00515

Resource Type:

Published Article

Journal Name:

Monthly Notices of the Royal Astronomical Society

Additional Journal Information:

Journal Name: Monthly Notices of the Royal Astronomical Society Journal Volume: 500 Journal Issue: 1; Journal ID: ISSN 0035-8711

Publisher:

Oxford University Press

Country of Publication:

United Kingdom

Language:

English