Dark energy survey year 3 results: likelihood-free, simulation-based <i>w</i>CDM inference with neural compression of weak-lensing map statistics

ABSTRACT We present simulation-based cosmological wcold dark matter (wCDM) inference using dark energy survey year 3 weak-lensing maps, via neural data compression of weak-lensing map summary statistics: power spectra, peak counts, and direct map-level compression/inference with convolutional neural networks (CNN). Using simulation-based inference, also known as likelihood-free or implicit inference, we use forward-modelled mock data to estimate posterior probability distributions of unknown parameters. This approach allows all statistical assumptions and uncertainties to be propagated through the forward-modelled mock data; these include sky masks, non-Gaussian shape noise, shape measurement bias, source galaxy clustering, photometric redshift uncertainty, intrinsic galaxy alignments, non-Gaussian density fields, neutrinos, and non-linear summary statistics. We include a series of tests to validate our inference results. This paper also describes the Gower Street simulation suite: 791 full-sky pkdgrav3 dark matter simulations, with cosmological model parameters sampled with a mixed active-learning strategy, from which we construct over 3000 mock dark energy survey lensing data sets. For wCDM inference, for which we allow $-1&amp;lt; w&amp;lt; -\frac{1}{3}$, our most constraining result uses power spectra combined with map-level (CNN) inference. Using gravitational lensing data only, this map-level combination gives $\Omega _{\rm m}= 0.283^{+0.020}_{-0.027}$, ${S_8 = 0.804^{+0.025}_{-0.017}}$, and $w &amp;lt; -0.80$ (with a 68 per cent credible interval); compared to the power spectrum inference, this is more than a factor of two improvement in dark energy parameter ($\Omega _{\rm DE}, w$) precision.

Authors

Institutions

Publication Details

Journal
Monthly Notices of the Royal Astronomical Society
Published
2024-11-27
DOI
https://doi.org/10.1093/mnras/stae2629
Citations
30
Primary Topic
Galaxies: Formation, Evolution, Phenomena
Type
article
Field-Weighted Citation Impact
9.95

Funders

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

Dark energy survey year 3 results: likelihood-free, simulation-based wCDM inference with neural compression of weak-lensing map statistics

A. Pieres, A. Roodman, Ting-Yun Cheng, I. Ferrero et al.
30 citations
Monthly Notices of the Royal Astronomical Society
Galaxies: Formation, Evolution, Phenomena
9.95
article

Dark energy survey year 3 results: likelihood-free, simulation-based wCDM inference with neural compression of weak-lensing map statistics

A. Pieres, A. Roodman, Ting-Yun Cheng, I. Ferrero, S. R. Hinton, M. Gatti, J. García-Bellido, S. Bocquet, S. Allam, L. N. da Costa, M. R. Becker, G. Giannini, R Chen, D. Gruen, Pablo Lemos, Bhuvnesh Jain, Gregory Tarlé, E. Suchyta, J. Prat, A. Amon, K. Eckert, A. Porredon, A Navarro-Alsina, J. Williamson, J. Zuntz, O. Alves, P. Wiseman, I. Tutusaus, G. M. Bernstein, C. Chang, David J. Brooks, C. Tucker, T. Shin, A Drlica-Wagner, D. Sanchez Cid, J. Mccullough, Eusebio Sanchez, J. Weller, N Jeffrey, J. Mena-Fernández, A. Campos, J. De Vicente, ENRIQUE GAZTAÑAGA, E. Sheldon, Marco Raveri, Tomasz Kacprzak, A. Ferté, D L Hollowood, Mike Jarvis, Joseph DeRose, C. P. Gutiérrez, C. Sánchez, S. Lee, T. M. Davis, K. Herner, Shivam Pandey, J Myles, M. E. C. Swanson, A. Choi, Justin Alsing, A. Alarcon, R. P. Rollins, H. T. Diehl, Dragan Huterer A, L Whiteway, C. Doux, J. L. Marshall, M. Yamamoto, A A Plazas Malagón, R A Gruendl, L F Secco, A. Carnero Rosell, D J James, T N Varga, M. C. Smith, E S Rykoff, B Yanny, Klaus Honscheid, M. Aguena, B. Yin, D. Bacon, S. Desai, I Sevilla-Noarbe, M. Sako, N. Weaverdyck, S Everett, J Frieman, O Lahav, M A Troxel, R Miquel, K Bechtol
article en
30 citations

Abstract

ABSTRACT We present simulation-based cosmological wcold dark matter (wCDM) inference using dark energy survey year 3 weak-lensing maps, via neural data compression of weak-lensing map summary statistics: power spectra, peak counts, and direct map-level compression/inference with convolutional neural networks (CNN). Using simulation-based inference, also known as likelihood-free or implicit inference, we use forward-modelled mock data to estimate posterior probability distributions of unknown parameters. This approach allows all statistical assumptions and uncertainties to be propagated through the forward-modelled mock data; these include sky masks, non-Gaussian shape noise, shape measurement bias, source galaxy clustering, photometric redshift uncertainty, intrinsic galaxy alignments, non-Gaussian density fields, neutrinos, and non-linear summary statistics. We include a series of tests to validate our inference results. This paper also describes the Gower Street simulation suite: 791 full-sky pkdgrav3 dark matter simulations, with cosmological model parameters sampled with a mixed active-learning strategy, from which we construct over 3000 mock dark energy survey lensing data sets. For wCDM inference, for which we allow $-1&amp;lt; w&amp;lt; -\frac{1}{3}$, our most constraining result uses power spectra combined with map-level (CNN) inference. Using gravitational lensing data only, this map-level combination gives $\Omega _{\rm m}= 0.283^{+0.020}_{-0.027}$, ${S_8 = 0.804^{+0.025}_{-0.017}}$, and $w &amp;lt; -0.80$ (with a 68 per cent credible interval); compared to the power spectrum inference, this is more than a factor of two improvement in dark energy parameter ($\Omega _{\rm DE}, w$) precision.

Monthly Notices of the Royal Astronomical SocietyVol. 536(2)
Institució Catalana de Recerca i Estudis Avançats (ES), Argonne National Laboratory (US), Oak Ridge National Laboratory (US), Centre National de la Recherche Scientifique (FR), Goddard Space Flight Center (US), Fermi National Accelerator Laboratory (US), Institute of Space Sciences (ES), Jet Propulsion Laboratory (US), Université Toulouse III - Paul Sabatier (FR), University of Wisconsin–Madison (US), Lawrence Berkeley National Laboratory (US), University of Illinois Urbana-Champaign (US), Universidad de La Laguna (ES), Stockholm University (SE), The University of Queensland (AU), Duke University (US), Université Fédérale de Toulouse Midi-Pyrénées (FR), Universidade Estadual de Campinas (UNICAMP) (BR), University of Oslo (NO), Institute for High Energy Physics (ES), Brookhaven National Laboratory (US), Princeton University (US), University of Cambridge (GB), University of Michigan (US), Mitchell Institute (US), SLAC National Accelerator Laboratory (US), University of Manchester (GB), ETH Zurich (CH), University of Chicago (US), Valongo Observatory (BR), Max Planck Institute for Extraterrestrial Physics (DE), Institut d'Estudis Espacials de Catalunya (ES), Kavli Institute for Particle Astrophysics and Cosmology (US), Excellence Cluster Origins (DE), Institute of Astronomy (RU), Center for Astrophysics Harvard & Smithsonian (US), National Observatory (BR), Istituto Nazionale di Fisica Nucleare, Sezione di Genova (IT), Instituto de Astrofísica de Canarias (ES), Institut National de Physique Nucléaire et de Physique des Particules (FR), National Center for Supercomputing Applications (US), Laboratoire de Physique Subatomique et de Cosmologie (FR), Wolfgang Pauli Institute (AT), Laboratório Interinstitucional de e-Astronomia (BR), Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas (ES), Institut de Recherche en Astrophysique et Planétologie (FR), University of Southampton (GB), Institute of Science and Technology, University College London (GB), UK Astronomy Technology Centre (GB), The Ohio State University (US), Stony Brook University (US), University of Portsmouth (GB), Universidad Autónoma de Madrid (ES), Indian Institute of Technology Hyderabad (IN), Carnegie Mellon University (US), University of Pennsylvania (US), Ludwig-Maximilians-Universität München (DE), KTH Royal Institute of Technology (SE), Université Grenoble Alpes (FR), Ruhr University Bochum (DE), Texas A&M University (US), Stanford University (US), University of Edinburgh (GB)
U.S. Department of Energy, University of California, Ohio State University, University of Chicago, National Centre for Supercomputing Applications, Higher Education Funding Council for England, National Energy Research Scientific Computing Center, Deutsche Forschungsgemeinschaft, Ministério da Ciência, Tecnologia e Inovação, Conselho Nacional de Desenvolvimento Científico e Tecnológico, Fundação Carlos Chagas Filho de Amparo à Pesquisa do Estado do Rio de Janeiro, Financiadora de Estudos e Projetos, University of Illinois at Urbana-Champaign, Office of Science, Horizon 2020 Framework Programme, Center for Cosmology and Astroparticle Physics, Ohio State University, Science and Technology Facilities Council, Centro de Investigaciones Energéticas, Medioambientales y Tecnológicas, Argonne National Laboratory, H2020 European Research Council
Affordable and clean energy
Openalex Percentile: Top 1%
Galaxies: Formation, Evolution, Phenomena
9.95
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.