Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables
In a novel approach employing implicit likelihood inference (ILI), also known as likelihood-free inference, we calibrate the parameters of cosmological hydrodynamic simulations against observations, which has previously been unfeasible due to the high computational cost of these simulations. For com...
| Main Authors: | , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2023-01-01
|
| Series: | The Astrophysical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4357/aca8fe |
| _version_ | 1797695627003428864 |
|---|---|
| author | Yongseok Jo Shy Genel Benjamin Wandelt Rachel S. Somerville Francisco Villaescusa-Navarro Greg L. Bryan Daniel Anglés-Alcázar Daniel Foreman-Mackey Dylan Nelson Ji-hoon Kim |
| author_facet | Yongseok Jo Shy Genel Benjamin Wandelt Rachel S. Somerville Francisco Villaescusa-Navarro Greg L. Bryan Daniel Anglés-Alcázar Daniel Foreman-Mackey Dylan Nelson Ji-hoon Kim |
| author_sort | Yongseok Jo |
| collection | DOAJ |
| description | In a novel approach employing implicit likelihood inference (ILI), also known as likelihood-free inference, we calibrate the parameters of cosmological hydrodynamic simulations against observations, which has previously been unfeasible due to the high computational cost of these simulations. For computational efficiency, we train neural networks as emulators on ∼1000 cosmological simulations from the CAMELS project to estimate simulated observables, taking as input the cosmological and astrophysical parameters, and use these emulators as surrogates for the cosmological simulations. Using the cosmic star formation rate density (SFRD) and, separately, the stellar mass functions (SMFs) at different redshifts, we perform ILI on selected cosmological and astrophysical parameters (Ω _m , σ _8 , stellar wind feedback, and kinetic black hole feedback) and obtain full six-dimensional posterior distributions. In the performance test, the ILI from the emulated SFRD (SMFs) can recover the target observables with a relative error of 0.17% (0.4%). We find that degeneracies exist between the parameters inferred from the emulated SFRD, confirmed with new full cosmological simulations. We also find that the SMFs can break the degeneracy in the SFRD, which indicates that the SMFs provide complementary constraints for the parameters. Further, we find that a parameter combination inferred from an observationally inferred SFRD reproduces the target observed SFRD very well, whereas, in the case of the SMFs, the inferred and observed SMFs show significant discrepancies that indicate potential limitations of the current galaxy formation modeling and calibration framework, and/or systematic differences and inconsistencies between observations of the SMFs. |
| first_indexed | 2024-03-12T03:15:01Z |
| format | Article |
| id | doaj.art-53d9e78069684f8080d82cadf087b435 |
| institution | Directory Open Access Journal |
| issn | 1538-4357 |
| language | English |
| last_indexed | 2024-03-12T03:15:01Z |
| publishDate | 2023-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal |
| spelling | doaj.art-53d9e78069684f8080d82cadf087b4352023-09-03T14:11:17ZengIOP PublishingThe Astrophysical Journal1538-43572023-01-0194416710.3847/1538-4357/aca8feCalibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth ObservablesYongseok Jo0https://orcid.org/0000-0003-3977-1761Shy Genel1https://orcid.org/0000-0002-3185-1540Benjamin Wandelt2https://orcid.org/0000-0002-5854-8269Rachel S. Somerville3Francisco Villaescusa-Navarro4https://orcid.org/0000-0002-4816-0455Greg L. Bryan5https://orcid.org/0000-0003-2630-9228Daniel Anglés-Alcázar6https://orcid.org/0000-0001-5769-4945Daniel Foreman-Mackey7https://orcid.org/0000-0002-9328-5652Dylan Nelson8https://orcid.org/0000-0001-8421-5890Ji-hoon Kim9https://orcid.org/0000-0003-4464-1160Center for Theoretical Physics, Department of Physics and Astronomy, Seoul National University , Seoul 08826, Republic of Korea ; kerex@snu.ac.krCenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USA; Columbia Astrophysics Laboratory, Columbia University , 550 West 120th Street, New York, NY, 10027, USACenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USA; Sorbonne Universite , CNRS, UMR 7095, Institut d’Astrophysique de Paris, 98 bis boulevard Arago, F-75014 Paris, FranceCenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USA; Department of Physics and Astronomy, Rutgers University , 136 Frelinghuysen Road, Piscataway, NJ 08854, USACenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USA; Department of Astrophysical Sciences, Princeton University , 4 Ivy Lane, Princeton, NJ 08544 USACenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USA; Department of Astronomy, Columbia University , 550 West 120th Street, New York, NY 10027, USACenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USA; Department of Physics, University of Connecticut , 196 Auditorium Road, U-3046, Storrs, CT 06269-3046, USACenter for Computational Astrophysics, Flatiron Institute , 162 5th Avenue, New York, NY, 10010, USAUniversität Heidelberg , Zentrum für Astronomie, Institut für Theoretische Astrophysik, Albert-Ueberle-Str. 2, D-69120 Heidelberg, GermanyCenter for Theoretical Physics, Department of Physics and Astronomy, Seoul National University , Seoul 08826, Republic of Korea ; kerex@snu.ac.krIn a novel approach employing implicit likelihood inference (ILI), also known as likelihood-free inference, we calibrate the parameters of cosmological hydrodynamic simulations against observations, which has previously been unfeasible due to the high computational cost of these simulations. For computational efficiency, we train neural networks as emulators on ∼1000 cosmological simulations from the CAMELS project to estimate simulated observables, taking as input the cosmological and astrophysical parameters, and use these emulators as surrogates for the cosmological simulations. Using the cosmic star formation rate density (SFRD) and, separately, the stellar mass functions (SMFs) at different redshifts, we perform ILI on selected cosmological and astrophysical parameters (Ω _m , σ _8 , stellar wind feedback, and kinetic black hole feedback) and obtain full six-dimensional posterior distributions. In the performance test, the ILI from the emulated SFRD (SMFs) can recover the target observables with a relative error of 0.17% (0.4%). We find that degeneracies exist between the parameters inferred from the emulated SFRD, confirmed with new full cosmological simulations. We also find that the SMFs can break the degeneracy in the SFRD, which indicates that the SMFs provide complementary constraints for the parameters. Further, we find that a parameter combination inferred from an observationally inferred SFRD reproduces the target observed SFRD very well, whereas, in the case of the SMFs, the inferred and observed SMFs show significant discrepancies that indicate potential limitations of the current galaxy formation modeling and calibration framework, and/or systematic differences and inconsistencies between observations of the SMFs.https://doi.org/10.3847/1538-4357/aca8feNonparametric inferenceLikelihood ratio testMagnetohydrodynamical simulationsCosmological parametersNeural networksBayes’ Theorem |
| spellingShingle | Yongseok Jo Shy Genel Benjamin Wandelt Rachel S. Somerville Francisco Villaescusa-Navarro Greg L. Bryan Daniel Anglés-Alcázar Daniel Foreman-Mackey Dylan Nelson Ji-hoon Kim Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables The Astrophysical Journal Nonparametric inference Likelihood ratio test Magnetohydrodynamical simulations Cosmological parameters Neural networks Bayes’ Theorem |
| title | Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables |
| title_full | Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables |
| title_fullStr | Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables |
| title_full_unstemmed | Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables |
| title_short | Calibrating Cosmological Simulations with Implicit Likelihood Inference Using Galaxy Growth Observables |
| title_sort | calibrating cosmological simulations with implicit likelihood inference using galaxy growth observables |
| topic | Nonparametric inference Likelihood ratio test Magnetohydrodynamical simulations Cosmological parameters Neural networks Bayes’ Theorem |
| url | https://doi.org/10.3847/1538-4357/aca8fe |
| work_keys_str_mv | AT yongseokjo calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT shygenel calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT benjaminwandelt calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT rachelssomerville calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT franciscovillaescusanavarro calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT greglbryan calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT danielanglesalcazar calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT danielforemanmackey calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT dylannelson calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables AT jihoonkim calibratingcosmologicalsimulationswithimplicitlikelihoodinferenceusinggalaxygrowthobservables |