pith. sign in

arxiv: 2606.11309 · v1 · pith:BE4BXFQCnew · submitted 2026-06-09 · 🌌 astro-ph.CO

Dark Energy Survey Year 3 results: optimized wCDM simulation-based inference with weak lensing map-level hybrid statistics

J. Williamson , T. L. Makinen , N. Porqueres , N. Jeffrey , A. Heavens , M. Gatti , B. D. Wandelt , L. Whiteway
show 80 more authors
J. Prat A. Alarcon A. Amon K. Bechtol M. R. Becker G. M. Bernstein A. Campos A. Carnero Rosell R. Chen A. Choi J. DeRose C. Doux A. Drlica-Wagner K. Eckert S. Everett A. Fert\'e Z. Gong D. Gruen R. A. Gruendl K. Herner M. Jarvis T. Kacprzak O. Lahav J. McCullough J. Myles A. Navarro-Alsina S. Pandey M. Raveri R. P. Rollins E. S. Rykoff C. S\'anchez L. F. Secco I. Sevilla-Noarbe E. Sheldon T. Shin A. Thomsen M. A. Troxel I. Tutusaus T. N. Varga B. Yanny B. Yin J. Zuntz T. M. C. Abbott M. Aguena F. Andrade-Oliveira D. Brooks R. Camilleri J. Carretero R. Cawthon M. Crocce L. N. da Costa T. M. Davis J. De Vicente S. Desai H. T. Diehl B. Flaugher J. Frieman J. Garc\'ia-Bellido G. Gutierrez S. R. Hinton D. L. Hollowood K. Kuehn J. L. Marshall J. Mena-Fern\'andez R. Miquel J. J. Mohr J. Muir A. A. Plazas Malag\'on A. Porredon A. Roodman E. Sanchez D. Sanchez Cid E. Suchyta M. E. C. Swanson C. To D. L. Tucker V. Vikram A. R. Walker N. Weaverdyck J. Weller
This is my paper

Pith reviewed 2026-06-27 11:56 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords weak gravitational lensingsimulation-based inferenceDark Energy Surveycosmological parametersdata compressionhybrid statisticswCDM cosmologymap-level analysis
0
0 comments X

The pith

DES Y3 weak lensing data alone yields the tightest joint constraints on Ωm, S8 and w from any survey to date.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper applies a Bayesian simulation-based inference pipeline to DES Year 3 weak lensing maps. It compresses the maps into seven summary statistics that combine the power spectrum with additional neural-network summaries chosen for maximum information content. The Gower Street simulations supply realistic forward models of the survey mask, photometric redshifts, intrinsic alignments, shear calibration, source clustering, shape noise and non-linear structure growth. This produces joint constraints on Ωm, S8 and w that improve the figure of merit by 60 percent over prior analyses and by nearly a factor of three over two-point statistics on the same data. A reader cares because the result tests the dark-energy equation of state using only weak lensing, without relying on galaxy clustering or other probes.

Core claim

Assuming a wCDM cosmology, the analysis yields S8 = 0.808 ± 0.017, Ωm = 0.325 ± 0.024 and w < -0.766 (68 per cent credible intervals). These are the most precise joint constraints on (Ωm, S8, w) from weak gravitational lensing data alone of any survey to date, improving the figure of merit for (Ωm, S8, w) by 60 per cent over the previous state-of-the-art and by almost a factor of 3 over two-point analyses of the same data.

What carries the argument

Hierarchical hybrid statistics that fuse the power spectrum with neural-based summaries, compressed via information theory to seven statistics and passed through a Bayesian simulation-based inference pipeline that uses the Gower Street simulations for forward modeling of systematics.

If this is right

  • The figure of merit for (Ωm, S8, w) improves by 60 per cent relative to the previous state-of-the-art analysis of the same data.
  • The constraints are almost three times tighter than those obtained from two-point statistics alone on DES Y3 weak lensing.
  • Coverage tests and checks against baryonic feedback confirm the robustness of the reported posteriors.
  • The same pipeline is intended for application to the forthcoming DES Y6 data set.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The reduction of map-level information to seven statistics indicates that most cosmological signal can be retained with extreme compression when the summaries are chosen by information-theoretic criteria.
  • If the Gower Street forward models remain accurate at higher precision, the hybrid method could be ported directly to other weak-lensing surveys to test consistency across data sets.
  • The framework's ability to marginalize over many systematics simultaneously suggests it could accommodate additional parameters such as neutrino mass or modified gravity without requiring new summary statistics.

Load-bearing premise

The Gower Street simulations accurately forward-model all major sources of systematic uncertainty and survey properties without residual biases that would shift the reported posteriors.

What would settle it

Re-running the inference on an independent set of simulations that incorporate additional baryonic feedback or altered photometric-redshift distributions and finding posterior means for S8 or w shifted by more than the reported 1-sigma uncertainties would falsify the forward-model accuracy.

Figures

Figures reproduced from arXiv: 2606.11309 by A. Alarcon, A. Amon, A. A. Plazas Malag\'on, A. Campos, A. Carnero Rosell, A. Choi, A. Drlica-Wagner, A. Fert\'e, A. Heavens, A. Navarro-Alsina, A. Porredon, A. Roodman, A. R. Walker, A. Thomsen, B. D. Wandelt, B. Flaugher, B. Yanny, B. Yin, C. Doux, C. S\'anchez, C. To, D. Brooks, D. Gruen, D. L. Hollowood, D. L. Tucker, D. Sanchez Cid, E. Sanchez, E. Sheldon, E. S. Rykoff, E. Suchyta, F. Andrade-Oliveira, G. Gutierrez, G. M. Bernstein, H. T. Diehl, I. Sevilla-Noarbe, I. Tutusaus, J. Carretero, J. DeRose, J. De Vicente, J. Frieman, J. Garc\'ia-Bellido, J. J. Mohr, J. L. Marshall, J. McCullough, J. Mena-Fern\'andez, J. Muir, J. Myles, J. Prat, J. Weller, J. Williamson, J. Zuntz, K. Bechtol, K. Eckert, K. Herner, K. Kuehn, L. F. Secco, L. N. da Costa, L. Whiteway, M. Aguena, M. A. Troxel, M. Crocce, M. E. C. Swanson, M. Gatti, M. Jarvis, M. Raveri, M. R. Becker, N. Jeffrey, N. Porqueres, N. Weaverdyck, O. Lahav, R. A. Gruendl, R. Camilleri, R. Cawthon, R. Chen, R. Miquel, R. P. Rollins, S. Desai, S. Everett, S. Pandey, S. R. Hinton, T. Kacprzak, T. L. Makinen, T. M. C. Abbott, T. M. Davis, T. N. Varga, T. Shin, V. Vikram, Z. Gong.

Figure 1
Figure 1. Figure 1: Plate diagram illustrating the hierarchical hybrid compression of data x with parameters 𝜽 and nuisance parameters 𝜂. The data are partitioned into subsets x0, x1, . . . , and the power spectrum 𝐶ℓ is computed from the full field. Compressions are performed sequentially: 𝐽 ∗ (𝐶ℓ ) compresses the power spectrum, then each 𝐹 ∗ (x𝑖 ) compresses a data subset conditional on all previously obtained summaries (i… view at source ↗
Figure 2
Figure 2. Figure 2: Hierarchical hybrid statistics scheme adapted for DES Y3 data products. The DES footprint is simulated over a simulation sampling distribution in cosmology 𝑝 sim ( 𝜃 ) with nuisance parameters (left). Empirical 𝐶ℓ vectors are computed from the full shear field (𝛾) footprint and compressed via MI maximization to dim(𝑡0 ) = 10 numbers. The reconstructed convergence field 𝜅 is split into patches A,B, and C an… view at source ↗
Figure 4
Figure 4. Figure 4: The DES Y3 full projected mass map at a HEALPix resolution of 1024. This convergence map is produced by Kaiser-Squires reconstruction, as detailed by Jeffrey & Gatti et al. (2021), using all the source tomographic bins from the metacalibration catalogue. and other observational conditions are below the statistical error budget, making the DES Y3 shape catalogue a key input for the suite of Y3 cosmology ana… view at source ↗
Figure 3
Figure 3. Figure 3: Samples from 𝑝(𝑛¯ (𝑧) | 𝑥phot) of the DES Y3 footprint redshift distribution from SOMPZ (Myles et al. 2021). only modelled in the initial conditions at a fixed mass of 0.06. The remaining runs beyond verification use the concept code to model the effect of neutrinos (Tram et al. 2019). 5.2 DES Y3 data The Dark Energy Survey (DES) is a multi-band imaging program carried out with the Dark Energy Camera on th… view at source ↗
Figure 5
Figure 5. Figure 5: Coverage test result (using TARP; Lemos et al. 2023) to validate the density estimation scheme for hybrid statistics. Using repeated mock data parameter inference, the fraction of true values in the appropriate credible intervals matches the expected fraction. The figure shows the result for all three patches and 𝐶ℓ compression. The shaded regions show 1- and 2-𝜎 contours (the standard deviation 𝜎 being ca… view at source ↗
Figure 6
Figure 6. Figure 6: Baryonic feedback systematic error test with the CosmoGridV1 simulations. The mean of the inferred marginal posterior distributions of mock data contaminated with baryonic feedback falls within the 0.3𝜎 of the uncontaminated marginal posterior distribution (this criterion derived from the standard DES Y3 test). form of systematic uncertainty at small scales in weak lensing. The training simulations are dar… view at source ↗
Figure 7
Figure 7. Figure 7: Three dimensional slices of hybrid summaries from simulations (scatter points) capture informative structure with respect to indicated parameter values (scatter colour gradient). The DES Y3 target data (red dot) falls squarely on this learned manifold, indicating that the compressed data is in-distribution with compressed simulations. This visual inspection is a blinded test, as it does not require assigni… view at source ↗
Figure 8
Figure 8. Figure 8: Modular posterior predictive distribution (PPD) test for the noise￾subtracted DES Y3 pseudo-𝐶ℓ power spectra. The shaded regions show the 1𝜎 and 2𝜎 credible intervals of the posterior predictive distribution. The DES Y3 measurements fall comfortably within the predicted distributions across all tomographic bin combinations. (0.3, 0.8), shown in [PITH_FULL_IMAGE:figures/full_fig_p012_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: Neural density estimator convergence test on Ωm and 𝑆8 for four (out of eight total) individual NDE ensemble members. This test was performed on the DES Y3 data by blindly shifting the mean value of the posteriors to fiducial values of Ωm = 0.3 and 𝑆8 = 0.8. 0.2 0.3 0.4 Ωm −0.9 −0.8 −0.7 −0.6 −0.5 −0.4 w 0.75 0.80 0.85 0.90 S8 0.75 0.80 0.85 0.90 S8 −0.8 −0.6 −0.4 w Cosmic Shear ξ± C`×CNN (Jeffrey et al.) … view at source ↗
Figure 10
Figure 10. Figure 10: Main result. DES Y3 𝑤CDM constraints: posterior probability distribution for parameters {Ωm, 𝑆8, 𝑤} with DES Y3 data. Hybrid statistics improves information extraction about all three parameters over two-point (Doux et al. 2022) and the existing CNN compression (Jeffrey et al. 2025). These constraints are consistent with Planck Collaboration (2020), the latter recalculated here with our analysis priors. w… view at source ↗
Figure 11
Figure 11. Figure 11: DES Y3 𝑤CDM results: we compare the joint posterior proba￾bility distributions between this work’s hybrid statistics and the results from Gatti et al. (2024a) which combines 2nd and 3rd order moments, scattering transforms and wavelet phase harmonics; Prat et al. (2026) which combines Betti numbers and 2nd order moments; and Jeffrey et al. (2025) which com￾bines the angular power spectrum with a field lev… view at source ↗
read the original abstract

We present cosmological constraints from the Dark Energy Survey Year 3 (DES Y3) weak lensing data using hierarchical hybrid statistics within a Bayesian simulation-based inference framework that is based on the Gower Street simulations. To maximize the precision of the inference, we have developed a new, information-theory based, data compression of the weak lensing maps to just seven highly informative summary statistics. The hybrid scheme exploits the high information content of the power spectrum, compressing both the power spectrum and neural-based summaries that are designed to extract further information. Our simulation-based approach enables principled forward modelling of all major sources of systematic uncertainty and survey properties into realistic mock observations, including the survey mask, photometric redshift uncertainties, intrinsic galaxy alignments, multiplicative shear calibration bias, source galaxy clustering, non-Gaussian shape noise, and non-linear structure formation. The summary statistics are then used in a Bayesian simulation-based inference pipeline. The inference is validated through coverage tests and checks for robustness against baryonic feedback. Assuming a $w$CDM cosmology, our analysis yields $S_8 = 0.808 \pm 0.017$, $\Omega_{\rm m} = 0.325 \pm 0.024$, and $w < -0.766$ (marginalized posterior 68 per cent credible intervals). This rigorous combination of information theory, physics- and neural network-based extreme data compression, and principled Bayesian analysis improves the figure of merit for $(\Omega_{\rm m}, S_8, w)$ by 60 per cent over the previous state-of-the-art, and by almost a factor of 3 over two-point analyses of the same data. They are the most precise joint constraints on $(\Omega_{\rm m}, S_8, w)$ from weak gravitational lensing data alone of any survey to date. We intend to apply this analysis to the more recent DES Y6 data.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The manuscript presents DES Y3 weak lensing constraints in wCDM using simulation-based inference on Gower Street mocks. It forward-models the survey mask, photo-z uncertainties, intrinsic alignments, multiplicative shear bias, source clustering, non-Gaussian shape noise and non-linear structure formation, compresses the maps to seven hybrid statistics (power spectrum plus neural-network summaries) via information-theoretic methods, and reports S8 = 0.808 ± 0.017, Ωm = 0.325 ± 0.024, w < -0.766 (68 % credible intervals). The analysis claims a 60 % figure-of-merit gain over prior state-of-the-art and a factor-of-three improvement over two-point statistics on the same data, positioning the result as the tightest joint (Ωm, S8, w) constraints from weak lensing alone.

Significance. If the forward-modeling accuracy and coverage tests are shown to bound residuals below the reported precision, the work would constitute a clear methodological advance: it demonstrates how principled extreme compression combined with full end-to-end simulation-based inference can extract substantially more cosmological information from existing weak-lensing maps than conventional two-point analyses.

major comments (2)
  1. [Abstract] Abstract: the coverage tests and baryonic-robustness checks are cited as validation, yet the text does not quantify whether these tests bound residual biases from the full list of modeled systematics (photometric redshift uncertainties, intrinsic alignments, source clustering, non-Gaussian shape noise) at the sub-σ level needed to support the quoted 0.017 uncertainty on S8. This directly affects the central claim that the reported credible intervals are unbiased.
  2. [Abstract] Abstract (hybrid statistics paragraph): the 60 % FoM improvement and factor-of-three gain over two-point analyses rest on the assumption that the seven compressed statistics inherit no residual modeling error from the Gower Street mocks; without an explicit propagation of simulation accuracy into the final posterior widths or a dedicated systematics-marginalization table, the improvement cannot be verified as free of forward-model bias.
minor comments (1)
  1. [Abstract] The abstract states that the inference is validated through coverage tests, but the manuscript would benefit from a dedicated subsection (or supplementary figure) showing the coverage probability as a function of the seven summary statistics rather than a single aggregate statement.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful and constructive review of our manuscript. The comments correctly identify areas where the abstract would benefit from greater explicitness regarding validation and error propagation. We address each point below and will make corresponding revisions to strengthen the presentation.

read point-by-point responses

  1. Referee: [Abstract] Abstract: the coverage tests and baryonic-robustness checks are cited as validation, yet the text does not quantify whether these tests bound residual biases from the full list of modeled systematics (photometric redshift uncertainties, intrinsic alignments, source clustering, non-Gaussian shape noise) at the sub-σ level needed to support the quoted 0.017 uncertainty on S8. This directly affects the central claim that the reported credible intervals are unbiased.

    Authors: We agree that the abstract would be improved by explicit quantification. The coverage tests in the full manuscript are end-to-end and incorporate all listed systematics (photo-z, IA, source clustering, shape noise, mask, shear calibration, and non-linear evolution). These tests recover input parameters with residuals well below the reported statistical precision. In the revised manuscript we will update the abstract to state that coverage tests bound maximum residual biases to <0.5σ on S8 and add a dedicated paragraph (or table) in the validation section summarizing the per-systematic residual bias levels from the mock suite. revision: yes

  2. Referee: [Abstract] Abstract (hybrid statistics paragraph): the 60 % FoM improvement and factor-of-three gain over two-point analyses rest on the assumption that the seven compressed statistics inherit no residual modeling error from the Gower Street mocks; without an explicit propagation of simulation accuracy into the final posterior widths or a dedicated systematics-marginalization table, the improvement cannot be verified as free of forward-model bias.

    Authors: The Gower Street simulations forward-model all relevant effects at the resolution required for DES Y3, and the compression is performed self-consistently within the same mocks. We nevertheless accept that an explicit propagation of any residual simulation inaccuracy would strengthen the claim. In revision we will add a short discussion of simulation accuracy and convergence tests together with a table that bounds or marginalizes the contribution of any remaining modeling error to the final posterior widths and to the reported FoM gains. revision: yes

Circularity Check

0 steps flagged

No significant circularity in the derivation chain

full rationale

The paper applies simulation-based Bayesian inference to DES Y3 weak lensing maps using Gower Street mocks that forward-model listed systematics, followed by information-theoretic compression to seven hybrid statistics and posterior sampling. The reported constraints (S8=0.808±0.017, Ωm=0.325±0.024, w<-0.766) and FoM gains are outputs of this external simulation-to-data comparison, validated by coverage tests; no equation or step reduces the final posteriors to a fitted parameter or self-citation by construction. The pipeline is self-contained against external benchmarks.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review supplies insufficient detail to enumerate specific free parameters, axioms, or invented entities; the wCDM model and accuracy of the Gower Street simulations are invoked but not expanded.

axioms (1)
  • domain assumption wCDM cosmology is the correct background model for the inference.
    Analysis is performed assuming wCDM as stated in the abstract.

pith-pipeline@v0.9.1-grok · 6381 in / 1362 out tokens · 36369 ms · 2026-06-27T11:56:42.360647+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

243 extracted references · 145 canonical work pages · 15 internal anchors

  1. [1]

    Optimal proposals for Approximate Bayesian Computation

    arXiv e-prints , pages =. doi:10.48550/arXiv.1808.06040 , keywords =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1808.06040

  2. [3]

    , volume = 156, number = 4, pages = 160, doi =

  3. [5]

    The Open Journal of Astrophysics , volume = 6, pages = 31, doi =

  4. [7]

    Living Reviews in Computational Astrophysics , volume = 8, number = 1, pages = 1, doi =

    Large-scale dark matter simulations , author =. Living Reviews in Computational Astrophysics , volume = 8, number = 1, pages = 1, doi =

  5. [8]

    , keywords =

    Numerical techniques for large cosmological N-body simulations , author =. , volume = 57, pages =. doi:10.1086/191003 , keywords =

    work page doi:10.1086/191003

  6. [9]

    , volume = 404, pages = 441, doi =

    Mapping the Dark Matter with Weak Gravitational Lensing , author =. , volume = 404, pages = 441, doi =

  7. [10]

    Galaxy lensing mocks from all-sky lensing maps , author =

    The MICE Grand Challenge light-cone simulation - III. Galaxy lensing mocks from all-sky lensing maps , author =. , volume = 447, number = 2, pages =. doi:10.1093/mnras/stu2464 , keywords =

    work page doi:10.1093/mnras/stu2464

  8. [14]

    Computational Astrophysics and Cosmology , volume = 4, number = 1, pages = 2, doi =

  9. [16]

    Advances in neural information processing systems , volume=

    Masked autoregressive flow for density estimation , author=. Advances in neural information processing systems , volume=. 2017 , doi =

    2017

  10. [17]

    Mixture density networks , author =

  11. [18]

    On Information and Sufficiency , author =. Ann. Math. Statist. , publisher =. doi:10.1214/aoms/1177729694 , url =

    work page doi:10.1214/aoms/1177729694

  12. [19]

    arXiv e-prints , pages =

    Variational Inference with Normalizing Flows , author =. arXiv e-prints , pages =

  13. [20]

    Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics , publisher =

    Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows , author =. Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics , publisher =

  14. [21]

    International Conference on Machine Learning , pages =

    Made: Masked autoencoder for distribution estimation , author =. International Conference on Machine Learning , pages =

  15. [22]

    Advances in neural information processing systems , pages =

    Improved variational inference with inverse autoregressive flow , author =. Advances in neural information processing systems , pages =

  16. [25]

    , volume = 477, number = 3, pages =

    Massive optimal data compression and density estimation for scalable, likelihood-free inference in cosmology , author =. , volume = 477, number = 3, pages =. doi:10.1093/mnras/sty819 , keywords =

    work page doi:10.1093/mnras/sty819

  17. [28]

    Deep Learning , author =

  18. [29]

    doi:10.48550/arXiv.2110.06581 , keywords =

    arXiv e-prints , pages =. doi:10.48550/arXiv.2110.06581 , keywords =

    work page doi:10.48550/arxiv.2110.06581

  19. [30]

    The Astrophysical Journal , publisher =

    A Framework for Obtaining Accurate Posteriors of Strong Gravitational Lensing Parameters with Flexible Priors and Implicit Likelihoods Using Density Estimation , author =. The Astrophysical Journal , publisher =. doi:10.3847/1538-4357/aca7c2 , url =

    work page doi:10.3847/1538-4357/aca7c2

  20. [31]

    Australian & New Zealand Journal of Statistics , publisher =

    Diagnostic tools for approximate Bayesian computation using the coverage property , author =. Australian & New Zealand Journal of Statistics , publisher =. 1301.3166 , archiveprefix =

    Pith/arXiv arXiv

  21. [33]

    doi:10.1051/0004-6361/202347236 , keywords =

    , volume = 679, pages =. doi:10.1051/0004-6361/202347236 , keywords =

    work page doi:10.1051/0004-6361/202347236

  22. [34]

    arXiv e-prints , pages =

    Hyper Suprime-Cam Year 3 Results: Cosmology from Cosmic Shear Two-point Correlation Functions , author =. arXiv e-prints , pages =. doi:10.48550/arXiv.2304.00702 , keywords =

    work page doi:10.48550/arxiv.2304.00702

  23. [37]

    , volume = 473, number = 2, pages =

    On the insufficiency of arbitrarily precise covariance matrices: non-Gaussian weak-lensing likelihoods , author =. , volume = 473, number = 2, pages =. doi:10.1093/mnras/stx2491 , keywords =

    work page doi:10.1093/mnras/stx2491

  24. [38]

    , volume = 477, number = 4, pages =

    The skewed weak lensing likelihood: why biases arise, despite data and theory being sound , author =. , volume = 477, number = 4, pages =. doi:10.1093/mnras/sty988 , keywords =

    work page doi:10.1093/mnras/sty988

  25. [39]

    , volume = 100, number = 2, pages =

    Cosmic shear: Inference from forward models , author =. , volume = 100, number = 2, pages =. doi:10.1103/PhysRevD.100.023519 , keywords =

    work page doi:10.1103/physrevd.100.023519

  26. [40]

    doi:10.1093/mnras/stw3161 , keywords =

    , volume = 466, number = 3, pages =. doi:10.1093/mnras/stw3161 , keywords =

    work page doi:10.1093/mnras/stw3161

  27. [41]

    New Journal of Physics , volume = 9, pages = 444, doi =

    Dark energy constraints from cosmic shear power spectra: impact of intrinsic alignments on photometric redshift requirements , author =. New Journal of Physics , volume = 9, pages = 444, doi =

  28. [42]

    doi:10.1103/PhysRevD.105.083518 , keywords =

    , volume = 105, number = 8, pages =. doi:10.1103/PhysRevD.105.083518 , keywords =

    work page doi:10.1103/physrevd.105.083518

  29. [43]

    Distinguishing standard and modified gravity cosmologies with machine learning , author =. Phys. Rev. D , publisher =. doi:10.1103/PhysRevD.100.023508 , url =

    work page doi:10.1103/physrevd.100.023508

  30. [44]

    Nature Astronomy , volume = 3, pages =

    An improved cosmological parameter inference scheme motivated by deep learning , author =. Nature Astronomy , volume = 3, pages =. doi:10.1038/s41550-018-0596-8 , keywords =

    work page doi:10.1038/s41550-018-0596-8

  31. [45]

    , volume = 474, number = 4, pages =

    KiDS-450 + 2dFLenS: Cosmological parameter constraints from weak gravitational lensing tomography and overlapping redshift-space galaxy clustering , author =. , volume = 474, number = 4, pages =. doi:10.1093/mnras/stx2820 , keywords =

    work page doi:10.1093/mnras/stx2820

  32. [47]

    , volume = 490, number = 2, pages =

    Weak lensing cosmology with convolutional neural networks on noisy data , author =. , volume = 490, number = 2, pages =. doi:10.1093/mnras/stz2610 , keywords =

    work page doi:10.1093/mnras/stz2610

  33. [49]

    , volume = 98, number = 12, pages = 123518, doi =

    Cosmological constraints from noisy convergence maps through deep learning , author =. , volume = 98, number = 12, pages = 123518, doi =

  34. [50]

    2000, title Efficient Computation of Cosmic Microwave Background Anisotropies in Closed Friedmann-Robertson-Walker Models , , 538, 473, 10.1086/309179

    , volume = 538, number = 2, pages =. doi:10.1086/309179 , keywords =

    work page doi:10.1086/309179

  35. [51]

    i\ \ Euclid\ \ /i\ \ preparation:

    M Knabenhans and J Stadel and D Potter and J Dakin and S Hannestad and T Tram and S Marelli and A Schneider and R Teyssier and P Fosalba and S Andreon and N Auricchio and C Baccigalupi and A Balaguera-Antol. i\ \ Euclid\ \ /i\ \ preparation:. , publisher =. doi:10.1093/mnras/stab1366 , url =

    work page doi:10.1093/mnras/stab1366

  36. [52]

    , volume = 150, pages = 150, doi =

    The Dark Energy Camera , author =. , volume = 150, pages = 150, doi =

  37. [53]

    arXiv e-prints , pages =

    Metacalibration: Direct Self-Calibration of Biases in Shear Measurement , author =. arXiv e-prints , pages =

  38. [54]

    , volume = 841, pages = 24, doi =

    Practical Weak-lensing Shear Measurement with Metacalibration , author =. , volume = 841, pages = 24, doi =

  39. [55]

    Meeting of the APS Division of Particles and Fields , collaboration =

    The Dark Energy Survey Data Management System , author =. Meeting of the APS Division of Particles and Fields , collaboration =

  40. [56]

    , volume = 130, number = 989, pages =

    The Dark Energy Survey Image Processing Pipeline , author =. , volume = 130, number = 989, pages =. doi:10.1088/1538-3873/aab4ef , keywords =

    work page doi:10.1088/1538-3873/aab4ef

  41. [58]

    , volume = 239, number = 2, pages = 18, doi =

  42. [59]

    , volume = 460, pages =

    The DES Science Verification weak lensing shear catalogues , author =. , volume = 460, pages =. doi:10.1093/mnras/stw990 , primaryclass =

    work page doi:10.1093/mnras/stw990

  43. [62]

    , volume = 100, number = 10, pages = 103506, doi =

  44. [66]

    doi:10.1093/mnras/stac2429 , keywords =

    , volume = 516, number = 4, pages =. doi:10.1093/mnras/stac2429 , keywords =

    work page doi:10.1093/mnras/stac2429

  45. [68]

    doi:10.48550/arXiv.2307.02545 , keywords =

    arXiv e-prints , pages =. doi:10.48550/arXiv.2307.02545 , keywords =

    work page doi:10.48550/arxiv.2307.02545

  46. [69]

    doi:10.1051/0004-6361/201834714 , keywords =

    , volume = 624, pages =. doi:10.1051/0004-6361/201834714 , keywords =

    work page doi:10.1051/0004-6361/201834714

  47. [70]

    http://www.w3.org/1998/Math/MathML

    Dark Energy Survey Year 3 results: Constraints on extensions to <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML" display="inline"><mml:mi mathvariant="normal">Λ</mml:mi><mml:mi>CDM</mml:mi></mml:math> with weak lensing and galaxy clustering , author =. Phys. Rev. D , publisher =. doi:10.1103/physrevd.107.083504 , issn =

    work page doi:10.1103/physrevd.107.083504 1998

  48. [71]

    doi:10.1093/mnras/stv1911 , keywords =

    , volume = 453, number = 4, pages =. doi:10.1093/mnras/stv1911 , keywords =

    work page doi:10.1093/mnras/stv1911

  49. [73]

    Approximating Likelihood Ratios with Calibrated Discriminative Classifiers

    Approximating Likelihood Ratios with Calibrated Discriminative Classifiers. arXiv e-prints , keywords =. doi:10.48550/arXiv.1506.02169 , archivePrefix =. 1506.02169 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1506.02169

  50. [74]

    and Salimans, Tim and Poole, Ben and Ho, Jonathan , title =

    Kingma, Diederik P. and Salimans, Tim and Poole, Ben and Ho, Jonathan , title =. Proceedings of the 35th International Conference on Neural Information Processing Systems , articleno =. 2024 , isbn =

    2024

  51. [76]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Measuring the Survey Transfer Function with Balrog. arXiv e-prints , keywords =

  52. [77]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Deep Field Optical + Near-Infrared Images and Catalogue. arXiv e-prints , keywords =

  53. [78]

    , keywords =

    Dark energy survey year 3 results: weak lensing shape catalogue. , keywords =. doi:10.1093/mnras/stab918 , primaryClass =

    work page doi:10.1093/mnras/stab918

  54. [79]

    , keywords =

    Dark Energy Survey year 3 results: point spread function modelling. , keywords =. doi:10.1093/mnras/staa3679 , primaryClass =

    work page doi:10.1093/mnras/staa3679

  55. [80]

    arXiv e-prints , keywords =

    DES Y3 results: Blending shear and redshift biases in image simulations. arXiv e-prints , keywords =

  56. [82]

    , keywords =

    Dark Energy Survey Year 3 Results: clustering redshifts - calibration of the weak lensing source redshift distributions with redMaGiC and BOSS/eBOSS. , keywords =. doi:10.1093/mnras/stab3311 , archivePrefix =. 2012.08569 , primaryClass =

    work page doi:10.1093/mnras/stab3311 2012

  57. [83]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Calibration of Lens Sample Redshift Distributions using Clustering Redshifts with BOSS/eBOSS. arXiv e-prints , keywords =

  58. [84]

    , keywords =

    Phenotypic redshifts with self-organizing maps: A novel method to characterize redshift distributions of source galaxies for weak lensing. , keywords =. doi:10.1093/mnras/stz2162 , primaryClass =

    work page doi:10.1093/mnras/stz2162

  59. [86]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Exploiting small-scale information with lensing shear ratios. arXiv e-prints , keywords =

  60. [87]

    arXiv e-prints , keywords =

    Assessing tension metrics with Dark Energy Survey and Planck data. arXiv e-prints , keywords =

  61. [88]

    , keywords =

    Dark energy survey internal consistency tests of the joint cosmological probes analysis with posterior predictive distributions. , keywords =. doi:10.1093/mnras/stab526 , primaryClass =

    work page doi:10.1093/mnras/stab526

  62. [89]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Covariance Modelling and its Impact on Parameter Estimation and Quality of Fit. arXiv e-prints , keywords =

  63. [90]

    , keywords =

    Blinding multiprobe cosmological experiments. , keywords =. doi:10.1093/mnras/staa965 , primaryClass =

    work page doi:10.1093/mnras/staa965

  64. [91]

    To be submitted to MNRAS

    Dark Energy Survey Year 3 Results: Multi-probe modeling validation. To be submitted to MNRAS. 2021

    2021

  65. [93]

    To be submitted to MNRAS

    Dark Energy Survey Year 3 Results: Galaxy clustering and systematics treatment for lens galaxy samples. To be submitted to MNRAS. 2021

    2021

  66. [94]

    To be submitted to MNRAS

    Dark Energy Survey Year 3 Results: Accurate measurement and modeling of galaxy-galaxy lensing. To be submitted to MNRAS. 2021

    2021

  67. [95]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Cosmology from Cosmic Shear and Robustness to Data Calibration. arXiv e-prints , keywords =

  68. [96]

    , keywords =

    Dark Energy Survey Year 3 results: Cosmology from cosmic shear and robustness to modeling uncertainty. , keywords =. doi:10.1103/PhysRevD.105.023515 , archivePrefix =. 2105.13544 , primaryClass =

    work page doi:10.1103/physrevd.105.023515

  69. [97]

    To be submitted to MNRAS

    Dark Energy Survey Year 3 Results: Constraints on cosmological parameters and galaxy bias models from galaxy clustering and galaxy-galaxy lensing. To be submitted to MNRAS. 2021

    2021

  70. [98]

    To be submitted to MNRAS

    Dark Energy Survey Year 3 Results: Measurement, modelling and cosmological impact of magnification in galaxy clustering and galaxy-galaxy lensing. To be submitted to MNRAS. 2021

    2021

  71. [99]

    , keywords =

    Dark Energy Survey Year 3 Results: Optimizing the Lens Sample in Combined Galaxy Clustering and Galaxy-Galaxy Lensing Analysis. , keywords =. doi:10.1103/PhysRevD.103.043503 , primaryClass =

    work page doi:10.1103/physrevd.103.043503

  72. [100]

    To be submitted to PRD

    Dark Energy Survey Year 3 Results: Cosmological Constraints from Galaxy Clustering and Galaxy-Galaxy Lensing using an Optimized Lens Sample. To be submitted to PRD. in prep

  73. [101]

    arXiv e-prints , keywords =

    Dark Energy Survey Year 3 Results: Cosmological Constraints from Galaxy Clustering and Weak Lensing. arXiv e-prints , keywords =

  74. [102]

    Audio Texture Synthesis with Scattering Moments

    Audio Texture Synthesis with Scattering Moments. arXiv e-prints , keywords =. doi:10.48550/arXiv.1311.0407 , archivePrefix =. 1311.0407 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1311.0407

  75. [103]

    Group Invariant Scattering

    Group Invariant Scattering. arXiv e-prints , keywords =. doi:10.48550/arXiv.1101.2286 , archivePrefix =. 1101.2286 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1101.2286

  76. [106]

    2006, Report of the Dark Energy Task Force ( @eprint arXiv astro-ph/0609591 ), https://arxiv.org/abs/astro-ph/0609591

    Albrecht A., et al. 2006, Report of the Dark Energy Task Force ( @eprint arXiv astro-ph/0609591 ), https://arxiv.org/abs/astro-ph/0609591

    Pith/arXiv arXiv 2006

  77. [109]

    Asgari M., others (KiDS Collaboration) 2025,

    2025

  78. [110]

    Asgari M., Tr \"o ster T., others (KiDS Collaboration) 2021, @doi [ ] 10.1051/0004-6361/202039070 , 645, A104

    work page doi:10.1051/0004-6361/202039070 2021

  79. [113]

    M., 1994, Technical Report NCRG/94/004, Mixture Density Networks, https://publications.aston.ac.uk/373/

    Bishop C. M., 1994, Technical Report NCRG/94/004, Mixture Density Networks, https://publications.aston.ac.uk/373/ . Neural Computing Research Group, Aston University, Birmingham, UK, https://publications.aston.ac.uk/373/

    1994

  80. [114]

    Bridle S., King L., 2007, @doi [New Journal of Physics] 10.1088/1367-2630/9/12/444 , http://adsabs.harvard.edu/abs/2007NJPh....9..444B 9, 444

    work page doi:10.1088/1367-2630/9/12/444 2007

Showing first 80 references.