Euclid preparation - LXXV. Estimating galaxy physical properties using CatBoost chained regressors with attention

Euclid will image ~14000 deg^2 of the extragalactic sky at visible and NIR wavelengths, providing a dataset of unprecedented size and richness that will facilitate a multitude of studies into the evolution of galaxies. In the vast majority of cases the main source of information will come from broad-band images and data products thereof. Therefore, there is a pressing need to identify or develop scalable yet reliable methodologies to estimate the redshift and physical properties of galaxies using broad-band photometry from Euclid, optionally including ground-based optical photometry also. To address this need, we present a novel method to estimate the redshift, stellar mass, star-formation rate, specific star-formation rate, E(B-V), and age of galaxies, using mock Euclid and ground-based photometry. The main novelty of our property-estimation pipeline is its use of the CatBoost implementation of gradient-boosted regression-trees, together with chained regression and an intelligent, automatic optimization of the training data. The pipeline also includes a computationally-efficient method to estimate prediction uncertainties, and, in the absence of ground-truth labels, provides accurate predictions for metrics of model performance up to z~2. We apply our pipeline to several datasets consisting of mock Euclid broad-band photometry and mock ground-based ugriz photometry, to evaluate the performance of our methodology for estimating the redshift and physical properties of galaxies detected in the Euclid Wide Survey. The quality of our photometric redshift and physical property estimates are highly competitive overall, validating our modeling approach. We find that the inclusion of ground-based optical photometry significantly improves the quality of the property estimation, highlighting the importance of combining Euclid data with ancillary ground-based optical data. (Abridged)

Authors

Institutions

Publication Details

Journal
ArXiv.org
Published
2025-01-01
DOI
https://doi.org/10.18154/rwth-2026-00946
Primary Topic
Astronomy and Astrophysical Research
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Euclid preparation - LXXV. Estimating galaxy physical properties using CatBoost chained regressors with attention

D Sapone, A Hall, G. Seidel, J. Brinchmann et al.
ArXiv.org
Astronomy and Astrophysical Research
article

Euclid preparation - LXXV. Estimating galaxy physical properties using CatBoost chained regressors with attention

D Sapone, A Hall, G. Seidel, J. Brinchmann, M. Martinelli, J. Hjorth, M. Ballardini, P. Tallada-Crespí, S Matthew, A Grazian, C. Sirignano, M. Brescia, I Tereno, E. Merlin, S Toft, I. Hook, I. Ferrero, A. Boucaud, M. Bolzonella, S. Borgani, O. Marggraf, K George, Patrice Simon, F. Courbin, Y. Copin, A. Díaz‐Sánchez, G. Gozaliasl, L Bisigello, Pierluigi Monaco, S. Bardelli, G. Desprez, L Patrizii, E Medinaceli, E. Munari, E. Sefusatti, G Verza, J Steinwagner, G Castignani, Claudio Moretti, O. Cucciati, V. Capobianco, L. Moscardini, T. Schrabback, S. Farrens, J. Odier, E. Zucca, S. Pires, D. Vergani, K Ganga, L. Gabarra, T. Vassallo, R. Scaramella, A Troja, Benjamin Joachimi, R. Toledo-Moreo, S Bruton, M. Moresco, M. Baes, R. Saglia, S. Farrens, M. Meneghetti, D. Bonino, F Fornari, F Grupp, A. Veropalumbo, M. Poncet, L. Valenziano, A. Cappi, A Kiessling, H. Hildebrandt, E Gaztanaga, E. Romelli, C. Burigana, I. Tutusaus, O. Mansutti, G. Zamorani, Matteo Viel, N. Martinet, I Risso, P Schneider, V. Allevato, J. Väliviita, S. Escoffier, C. Padilla Aranda, E. Maiorano, B. Gillis, M Sahlén, F. Dubath, S Casas, A. Hornstrup, F Pasian, A Calabrò, C. Tao, C. Tortora, M. Pöntinen, R. B. Metcalf, A. Secroun, M. Farina, F. Mannucci, L. Guzzo, V. Kansal, S Anselmi, Sergio Serrano, J. Weller, S. Ligori, R. Farinelli, F. Giacomini, S. Galeotta, S Paltani, M. Melchior, E. Franceschi, A Cimatti, K Jahnke, Marco Baldi, M Calabrese, T Castro, S. Dusini, N. A. Walton, G Cañas-Herrera, Alain Blanchard, K. Markovič, A Amara, S. Camera, Z. Sakr, L. Legrand, E. Bozzo, Romain Teyssier, M. Kilbinger, J. Dinis, V. Lindholm, R. Rébolo, M. Sereno, E. Keihänen, S. Cavuoti, N. Auricchio, M. Frailis, V Scottez, A. Spurio Mancini, G Polenta, R Massey, C. Carbone, S. Kermiche, Arthur Loureiro, G. Meylan, M. Scodeggio, S. Contarini, E. Rossetti, A Renzi, M. G. Castellano, E. Branchini, C. Porciani, A. Humphrey, L. Pozzetti, Y. Akrami, K. Jahnkę, A. Da Silva, Yang Wang, L. Maurin, G. Riccio, D Karagiannis, Ryota Nakajima, G de Lucia, V. Popa, M Kunz, F Torradeflot, Y. Mellier, G Maggio, D. Maino, X. Dupac, N Mauri, M. Tenti, J García-Bellido, V. Pettorino, A. Schneider, G Congedo, H. Degaudenzi, H. Kurki‐Suonio, M. Wiesmann, G. Testera, J Rhodes, F. Hormuth, G Morgante, F. Raison, K. Pedersen, J.A. Escartin Vigo, J. Zoubian, B. Kubik, G Mainetti, R Cabanac, I. Lloro, S. Andreon, L Conversi, W. Holmes, F Marulli, M Tucci, C Baccigalupi, S Mei, J Carretero, A Gregorio, J. Coupon, M Magliocchetti, C Valieri, C. Bodendorf, L. Stanco, S. Tosi, M. Crocce, M Roncarelli, A Biviano, C Colodro-Conde, S. Ferriol, M. Kümmel, D Potter, C.J Conselice
article en

Abstract

Euclid will image ~14000 deg^2 of the extragalactic sky at visible and NIR wavelengths, providing a dataset of unprecedented size and richness that will facilitate a multitude of studies into the evolution of galaxies. In the vast majority of cases the main source of information will come from broad-band images and data products thereof. Therefore, there is a pressing need to identify or develop scalable yet reliable methodologies to estimate the redshift and physical properties of galaxies using broad-band photometry from Euclid, optionally including ground-based optical photometry also. To address this need, we present a novel method to estimate the redshift, stellar mass, star-formation rate, specific star-formation rate, E(B-V), and age of galaxies, using mock Euclid and ground-based photometry. The main novelty of our property-estimation pipeline is its use of the CatBoost implementation of gradient-boosted regression-trees, together with chained regression and an intelligent, automatic optimization of the training data. The pipeline also includes a computationally-efficient method to estimate prediction uncertainties, and, in the absence of ground-truth labels, provides accurate predictions for metrics of model performance up to z~2. We apply our pipeline to several datasets consisting of mock Euclid broad-band photometry and mock ground-based ugriz photometry, to evaluate the performance of our methodology for estimating the redshift and physical properties of galaxies detected in the Euclid Wide Survey. The quality of our photometric redshift and physical property estimates are highly competitive overall, validating our modeling approach. We find that the inclusion of ground-based optical photometry significantly improves the quality of the property estimation, highlighting the importance of combining Euclid data with ancillary ground-based optical data. (Abridged)

ArXiv.org
Université Paris Cité (FR), Université de Lille (FR), Commissariat à l'Énergie Atomique et aux Énergies Alternatives (FR), Centre National d'Études Spatiales (FR), Institut d'Astrophysique de Paris (FR), European Space Agency (FR), Laboratoire AstroParticule et Cosmologie (FR), Astrophysique, Instrumentation et Modélisation (FR), Janssen (France) (FR), Laboratoire Univers et Théories (FR), Centre de physique des particules de Marseille (FR), Laboratoire d’Astrophysique de Marseille (FR), Institut d'Astrophysique Spatiale (FR), Lagrange Laboratory (FR), Château Gombert (FR), CEA Paris-Saclay (FR), Laboratoire de Physique Subatomique et de Cosmologie (FR), Centre de Calcul de l’Institut National de Physique Nucléaire et de Physique des Particules (FR), Laboratoire Paul Painlevé (FR), Laboratoire de Physique des Plasmas (FR), Institute of Nuclear Physics of Lyon (FR), Institut de Recherche en Astrophysique et Planétologie (FR), Institut de Physique des 2 Infinis de Lyon (FR)
National Aeronautics and Space Administration, Nvidia, European Space Agency, Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Magyar Tudományos Akadémia, Agenzia Spaziale Italiana, National Astronomical Observatory of Japan, Agenția Spațială Română, Norsk Romsenter, Fundação para a Ciência e a Tecnologia
Openalex Percentile: Top 71%
Astronomy and Astrophysical Research
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.