TY - JOUR
T1 - Robust Field-level Likelihood-free Inference with Galaxies
AU - de Santi, Natalí S.M.
AU - Shao, Helen
AU - Villaescusa-Navarro, Francisco
AU - Abramo, L. Raul
AU - Teyssier, Romain
AU - Villanueva-Domingo, Pablo
AU - Ni, Yueying
AU - Anglés-Alcázar, Daniel
AU - Genel, Shy
AU - Hernández-Martínez, Elena
AU - Steinwandel, Ulrich P.
AU - Lovell, Christopher C.
AU - Dolag, Klaus
AU - Castro, Tiago
AU - Vogelsberger, Mark
N1 - Publisher Copyright:
© 2023. The Author(s). Published by the American Astronomical Society.
PY - 2023/7/1
Y1 - 2023/7/1
N2 - We train graph neural networks to perform field-level likelihood-free inference using galaxy catalogs from state-of-the-art hydrodynamic simulations of the CAMELS project. Our models are rotational, translational, and permutation invariant and do not impose any cut on scale. From galaxy catalogs that only contain 3D positions and radial velocities of ∼1000 galaxies in tiny ( 25 h − 1 Mpc ) 3 volumes our models can infer the value of Ωm with approximately 12% precision. More importantly, by testing the models on galaxy catalogs from thousands of hydrodynamic simulations, each having a different efficiency of supernova and active galactic nucleus feedback, run with five different codes and subgrid models—IllustrisTNG, SIMBA, Astrid, Magneticum, SWIFT-EAGLE—we find that our models are robust to changes in astrophysics, subgrid physics, and subhalo/galaxy finder. Furthermore, we test our models on 1024 simulations that cover a vast region in parameter space—variations in five cosmological and 23 astrophysical parameters—finding that the model extrapolates really well. Our results indicate that the key to building a robust model is the use of both galaxy positions and velocities, suggesting that the network has likely learned an underlying physical relation that does not depend on galaxy formation and is valid on scales larger than ∼10 h −1 kpc.
AB - We train graph neural networks to perform field-level likelihood-free inference using galaxy catalogs from state-of-the-art hydrodynamic simulations of the CAMELS project. Our models are rotational, translational, and permutation invariant and do not impose any cut on scale. From galaxy catalogs that only contain 3D positions and radial velocities of ∼1000 galaxies in tiny ( 25 h − 1 Mpc ) 3 volumes our models can infer the value of Ωm with approximately 12% precision. More importantly, by testing the models on galaxy catalogs from thousands of hydrodynamic simulations, each having a different efficiency of supernova and active galactic nucleus feedback, run with five different codes and subgrid models—IllustrisTNG, SIMBA, Astrid, Magneticum, SWIFT-EAGLE—we find that our models are robust to changes in astrophysics, subgrid physics, and subhalo/galaxy finder. Furthermore, we test our models on 1024 simulations that cover a vast region in parameter space—variations in five cosmological and 23 astrophysical parameters—finding that the model extrapolates really well. Our results indicate that the key to building a robust model is the use of both galaxy positions and velocities, suggesting that the network has likely learned an underlying physical relation that does not depend on galaxy formation and is valid on scales larger than ∼10 h −1 kpc.
UR - http://www.scopus.com/inward/record.url?scp=85165671541&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85165671541&partnerID=8YFLogxK
U2 - 10.3847/1538-4357/acd1e2
DO - 10.3847/1538-4357/acd1e2
M3 - Article
AN - SCOPUS:85165671541
SN - 0004-637X
VL - 952
JO - Astrophysical Journal
JF - Astrophysical Journal
IS - 1
M1 - 69
ER -