The CAMELS Multifield Data Set: Learning the Universe's Fundamental Parameters with Artificial Intelligence

Francisco Villaescusa-Navarro; Shy Genel; Daniel Anglés-Alcázar; Leander Thiele; Romeel Dave; Desika Narayanan; Andrina Nicola; Yin Li; Pablo Villanueva-Domingo; Benjamin Wandelt; David N. Spergel; Rachel S. Somerville; Jose Manuel Zorrilla Matilla; Faizan G. Mohammad; Sultan Hassan; Helen Shao; Digvijay Wadekar; Michael Eickenberg; Kaze W.K. Wong; Gabriella Contardo; Yongseok Jo; Emily Moser; Erwin T. Lau; Luis Fernando Machado Poletti Valle; Lucia A. Perez; Daisuke Nagai; Nicholas Battaglia; Mark Vogelsberger

doi:10.3847/1538-4365/ac5ab0

The CAMELS Multifield Data Set: Learning the Universe's Fundamental Parameters with Artificial Intelligence

Francisco Villaescusa-Navarro, Shy Genel, Daniel Anglés-Alcázar, Leander Thiele, Romeel Dave, Desika Narayanan, Andrina Nicola, Yin Li, Pablo Villanueva-Domingo, Benjamin Wandelt, David N. Spergel, Rachel S. Somerville, Jose Manuel Zorrilla Matilla, Faizan G. Mohammad, Sultan Hassan, Helen Shao, Digvijay Wadekar, Michael Eickenberg, Kaze W.K. Wong, Gabriella ContardoYongseok Jo, Emily Moser, Erwin T. Lau, Luis Fernando Machado Poletti Valle, Lucia A. Perez, Daisuke Nagai, Nicholas Battaglia, Mark Vogelsberger

Physics

Research output: Contribution to journal › Article › peer-review

Abstract

We present the Cosmology and Astrophysics with Machine Learning Simulations (CAMELS) Multifield Data set (CMD), a collection of hundreds of thousands of 2D maps and 3D grids containing many different properties of cosmic gas, dark matter, and stars from more than 2000 distinct simulated universes at several cosmic times. The 2D maps and 3D grids represent cosmic regions that span ∼100 million light-years and have been generated from thousands of state-of-the-art hydrodynamic and gravity-only N-body simulations from the CAMELS project. Designed to train machine-learning models, CMD is the largest data set of its kind containing more than 70 TB of data. In this paper we describe CMD in detail and outline a few of its applications. We focus our attention on one such task, parameter inference, formulating the problems we face as a challenge to the community. We release all data and provide further technical details at https://camels-multifield-dataset.readthedocs.io.

Original language	English (US)
Article number	61
Journal	Astrophysical Journal, Supplement Series
Volume	259
Issue number	2
DOIs	https://doi.org/10.3847/1538-4365/ac5ab0
State	Published - Apr 1 2022

ASJC Scopus subject areas

Astronomy and Astrophysics
Space and Planetary Science

Access to Document

10.3847/1538-4365/ac5ab0

Cite this

Villaescusa-Navarro, F., Genel, S., Anglés-Alcázar, D., Thiele, L., Dave, R., Narayanan, D., Nicola, A., Li, Y., Villanueva-Domingo, P., Wandelt, B., Spergel, D. N., Somerville, R. S., Zorrilla Matilla, J. M., Mohammad, F. G., Hassan, S., Shao, H., Wadekar, D., Eickenberg, M., Wong, K. W. K., ... Vogelsberger, M. (2022). The CAMELS Multifield Data Set: Learning the Universe's Fundamental Parameters with Artificial Intelligence. Astrophysical Journal, Supplement Series, 259(2), Article 61. https://doi.org/10.3847/1538-4365/ac5ab0

Villaescusa-Navarro, F, Genel, S, Anglés-Alcázar, D, Thiele, L, Dave, R, Narayanan, D, Nicola, A, Li, Y, Villanueva-Domingo, P, Wandelt, B, Spergel, DN, Somerville, RS, Zorrilla Matilla, JM, Mohammad, FG, Hassan, S, Shao, H, Wadekar, D, Eickenberg, M, Wong, KWK, Contardo, G, Jo, Y, Moser, E, Lau, ET, Machado Poletti Valle, LF, Perez, LA, Nagai, D, Battaglia, N & Vogelsberger, M 2022, 'The CAMELS Multifield Data Set: Learning the Universe's Fundamental Parameters with Artificial Intelligence', Astrophysical Journal, Supplement Series, vol. 259, no. 2, 61. https://doi.org/10.3847/1538-4365/ac5ab0

@article{3289e004421548709804a6aeb1185a74,

title = "The CAMELS Multifield Data Set: Learning the Universe's Fundamental Parameters with Artificial Intelligence",

abstract = "We present the Cosmology and Astrophysics with Machine Learning Simulations (CAMELS) Multifield Data set (CMD), a collection of hundreds of thousands of 2D maps and 3D grids containing many different properties of cosmic gas, dark matter, and stars from more than 2000 distinct simulated universes at several cosmic times. The 2D maps and 3D grids represent cosmic regions that span ∼100 million light-years and have been generated from thousands of state-of-the-art hydrodynamic and gravity-only N-body simulations from the CAMELS project. Designed to train machine-learning models, CMD is the largest data set of its kind containing more than 70 TB of data. In this paper we describe CMD in detail and outline a few of its applications. We focus our attention on one such task, parameter inference, formulating the problems we face as a challenge to the community. We release all data and provide further technical details at https://camels-multifield-dataset.readthedocs.io.",

author = "Francisco Villaescusa-Navarro and Shy Genel and Daniel Angl{\'e}s-Alc{\'a}zar and Leander Thiele and Romeel Dave and Desika Narayanan and Andrina Nicola and Yin Li and Pablo Villanueva-Domingo and Benjamin Wandelt and Spergel, {David N.} and Somerville, {Rachel S.} and {Zorrilla Matilla}, {Jose Manuel} and Mohammad, {Faizan G.} and Sultan Hassan and Helen Shao and Digvijay Wadekar and Michael Eickenberg and Wong, {Kaze W.K.} and Gabriella Contardo and Yongseok Jo and Emily Moser and Lau, {Erwin T.} and {Machado Poletti Valle}, {Luis Fernando} and Perez, {Lucia A.} and Daisuke Nagai and Nicholas Battaglia and Mark Vogelsberger",

note = "Publisher Copyright: {\textcopyright} 2022. The Author(s). Published by the American Astronomical Society.",

year = "2022",

month = apr,

day = "1",

doi = "10.3847/1538-4365/ac5ab0",

language = "English (US)",

volume = "259",

journal = "Astrophysical Journal, Supplement Series",

issn = "0067-0049",

publisher = "American Astronomical Society",

number = "2",

}

TY - JOUR

T1 - The CAMELS Multifield Data Set

T2 - Learning the Universe's Fundamental Parameters with Artificial Intelligence

AU - Villaescusa-Navarro, Francisco

AU - Genel, Shy

AU - Anglés-Alcázar, Daniel

AU - Thiele, Leander

AU - Dave, Romeel

AU - Narayanan, Desika

AU - Nicola, Andrina

AU - Li, Yin

AU - Villanueva-Domingo, Pablo

AU - Wandelt, Benjamin

AU - Spergel, David N.

AU - Somerville, Rachel S.

AU - Zorrilla Matilla, Jose Manuel

AU - Mohammad, Faizan G.

AU - Hassan, Sultan

AU - Shao, Helen

AU - Wadekar, Digvijay

AU - Eickenberg, Michael

AU - Wong, Kaze W.K.

AU - Contardo, Gabriella

AU - Jo, Yongseok

AU - Moser, Emily

AU - Lau, Erwin T.

AU - Machado Poletti Valle, Luis Fernando

AU - Perez, Lucia A.

AU - Nagai, Daisuke

AU - Battaglia, Nicholas

AU - Vogelsberger, Mark

PY - 2022/4/1

Y1 - 2022/4/1

N2 - We present the Cosmology and Astrophysics with Machine Learning Simulations (CAMELS) Multifield Data set (CMD), a collection of hundreds of thousands of 2D maps and 3D grids containing many different properties of cosmic gas, dark matter, and stars from more than 2000 distinct simulated universes at several cosmic times. The 2D maps and 3D grids represent cosmic regions that span ∼100 million light-years and have been generated from thousands of state-of-the-art hydrodynamic and gravity-only N-body simulations from the CAMELS project. Designed to train machine-learning models, CMD is the largest data set of its kind containing more than 70 TB of data. In this paper we describe CMD in detail and outline a few of its applications. We focus our attention on one such task, parameter inference, formulating the problems we face as a challenge to the community. We release all data and provide further technical details at https://camels-multifield-dataset.readthedocs.io.

AB - We present the Cosmology and Astrophysics with Machine Learning Simulations (CAMELS) Multifield Data set (CMD), a collection of hundreds of thousands of 2D maps and 3D grids containing many different properties of cosmic gas, dark matter, and stars from more than 2000 distinct simulated universes at several cosmic times. The 2D maps and 3D grids represent cosmic regions that span ∼100 million light-years and have been generated from thousands of state-of-the-art hydrodynamic and gravity-only N-body simulations from the CAMELS project. Designed to train machine-learning models, CMD is the largest data set of its kind containing more than 70 TB of data. In this paper we describe CMD in detail and outline a few of its applications. We focus our attention on one such task, parameter inference, formulating the problems we face as a challenge to the community. We release all data and provide further technical details at https://camels-multifield-dataset.readthedocs.io.

UR - http://www.scopus.com/inward/record.url?scp=85129076827&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85129076827&partnerID=8YFLogxK

U2 - 10.3847/1538-4365/ac5ab0

DO - 10.3847/1538-4365/ac5ab0

M3 - Article

AN - SCOPUS:85129076827

SN - 0067-0049

VL - 259

JO - Astrophysical Journal, Supplement Series

JF - Astrophysical Journal, Supplement Series

IS - 2

M1 - 61

ER -

The CAMELS Multifield Data Set: Learning the Universe's Fundamental Parameters with Artificial Intelligence

Abstract

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this