pith. sign in

arxiv: 2606.12255 · v1 · pith:GWZG3J27new · submitted 2026-06-10 · 🌌 astro-ph.CO

Towards Practical Field-Level Inference for Weak Lensing

Yuuki Omori , Justine Zeghal , Chihway Chang , Fran\c{c}ois Lanusse , Laurence Perreault-Levasseur This is my paper

Pith reviewed 2026-06-27 08:47 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords field-level inferenceweak lensingcosmological constraintsforward modelingpower spectrumparticle-mesh simulationLagrangian perturbation theorynonlinear structure growth
0
0 comments X

The pith

Field-level inference from weak lensing maps extracts significantly more cosmological information than power-spectrum methods alone.

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

The paper tests whether directly comparing full observed weak lensing maps to simulated maps can tighten constraints on cosmological parameters beyond what two-point statistics achieve. Using identical forward-modeling pipelines based on Lagrangian perturbation theory and particle-mesh N-body simulations, the authors run both implicit and explicit inference on the same data and find consistent gains, especially when small scales enter the particle-mesh models. A sympathetic reader cares because upcoming surveys will measure higher-order correlations with high precision, and any method that captures more of that information reduces uncertainty on quantities such as the matter density and dark-energy equation of state. Coverage tests confirm the implicit analyses are well calibrated, supporting the reliability of the reported improvements.

Core claim

Field-level inference using 8-million-parameter forward models based on Lagrangian perturbation theory and particle-mesh N-body evolution produces posteriors that agree between implicit and explicit methods and deliver significant gains in cosmological information relative to power-spectrum analyses performed with the same pipeline, with the largest improvements appearing when small scales are retained in the particle-mesh models.

What carries the argument

Field-level inference, the direct or summary-based comparison of observed weak lensing maps to forward-modeled maps generated from the same cosmological parameters.

If this is right

  • Including small scales in the particle-mesh forward model increases the information gain over power spectra more than Lagrangian perturbation theory models do.
  • Implicit and explicit field-level methods produce nearly identical posteriors when applied to the same simulated maps.
  • Coverage tests on the implicit analyses confirm that the recovered posteriors are statistically well calibrated.
  • Remaining modeling and computational challenges must be solved before particle-mesh-based explicit field-level inference can be used on real observations.

Where Pith is reading between the lines

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

  • The demonstrated gains suggest that map-level methods could help resolve parameter tensions if the forward models are later extended to include baryonic physics.
  • Similar field-level comparisons could be applied to other two-point-limited probes such as galaxy clustering to test whether the information gain generalizes.
  • If the calibration holds on real data, the approach would allow tighter joint constraints when combined with other cosmological datasets without double-counting two-point information.

Load-bearing premise

The chosen forward models capture enough of the true nonlinear features in real weak lensing maps that differences between map-level and power-spectrum constraints reflect genuine extra information rather than model mismatch.

What would settle it

Running the same field-level pipeline on actual survey data and obtaining cosmological constraints that are statistically consistent with power-spectrum results from the same data, or that disagree with independent probes at a level larger than expected from the reported gains.

Figures

Figures reproduced from arXiv: 2606.12255 by Chihway Chang, Fran\c{c}ois Lanusse, Justine Zeghal, Laurence Perreault-Levasseur, Yuuki Omori.

Figure 1
Figure 1. Figure 1: — Redshift distribution used in this work. The solid filled curves represent the source n(z) at z = 0.7 and z = 0.8, while the dashed lines represent the lensing kernels. the cosmology dependence of the redshift-to-distance re￾lation: across the prior range, the lensing kernel must re￾main fully contained within the simulated volume. The narrow redshift bins used here are therefore not intended to represen… view at source ↗
Figure 2
Figure 2. Figure 2: — Comparison of the noiseless convergence maps (second tomographic bin) between LPT and PM simulations prior to adding noise and applying the ℓmax cut. 0 200 400 600 800 1000 ` 10−11 10−10 10−9 10−8 10−7 10−6 Cκκ ` Analytic nonlinear Analytic linear Mean LPT zmid = 0.7 Mean LPT zmid = 0.8 Mean PM zmid = 0.7 Mean PM zmid = 0.8 [PITH_FULL_IMAGE:figures/full_fig_p007_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: — Comparison of the mean recovered power spectra for the two source bins, plotted with the scatter from 2000 noiseless realizations. The solid and dashed lines show the analytical calculations using the same source-galaxy n(z) and the nonlinear and linear matter power spectra, respectively. The dotted line shows an effective model in which the analytic nonlinear prediction is convolved with a 5′ beam. The … view at source ↗
Figure 4
Figure 4. Figure 4: — Cosmological constraints obtained using power-spectrum, implicit, and explicit inference with ℓmax = 400 and ℓmax = 1000 for LPT (upper panels) and PM (lower panels). Within each panel, all three approaches are applied to the same data realization. Note that the vertical axis is S8 = σ8(Ωm/0.3)0.5 , not S lin 8 = σ8(Ωm/0.3)α, on which the inference is run. with PI4ff/PI10ff. The 2D posteriors are broadly… view at source ↗
Figure 5
Figure 5. Figure 5: — TARP calibration for posterior uncertainty. Each panel shows the empirical expected coverage as a function of nominal credibility level for the configuration labeled in the legend. The dashed diagonal line indicates perfect calibration (expected coverage equals credibility); deviations below (above) the line indicate overconfident (underconfident) posteriors. Solid curves show the mean over simulations, … view at source ↗
Figure 6
Figure 6. Figure 6: — Comparison of implicit field-level constraints obtained using the baseline ResNet compressor and a simpler CNN compressor. Both architectures recover similar degeneracy directions, but the ResNet compressor gives consistently tighter constraints. block, Λb, contains the rest of the latent field and is kept diagonal, with entries given by the analytic Fourier-space inverse mass. After fixing this block st… view at source ↗
read the original abstract

Nonlinear structure growth generates higher-order correlations and morphological features in the cosmic density field that cannot be fully characterized by two-point statistics. Upcoming surveys will measure these features with greater precision, making it essential to develop methods capable of extracting as much cosmological information as possible from them. Field-level inference (FLI) is one such approach, in which cosmological parameters are constrained by comparing observed maps to forward-modeled maps, either directly or through learned summaries that retain map-level information. In this work, we compare FLI with power-spectrum-based inference using the same forward-modeling pipeline for generating weak lensing maps, with the goal of quantifying the gain from map-level analysis relative to two-point statistics. We perform this comparison with both implicit and explicit inference methods, using 8-million-parameter forward models based on Lagrangian perturbation theory and particle-mesh (PM) N-body evolution. The two FLI approaches yield closely consistent posteriors; this agreement, together with coverage tests confirming the calibration of the implicit analyses, gives us confidence in the recovered field-level constraints. Relative to the power-spectrum-based analyses, these results show significant gains in cosmological information, especially when small scales are included in the PM-based forward model. We then discuss the remaining challenges that must be addressed before PM-based explicit FLI can be applied to observational datasets.

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

0 major / 2 minor

Summary. The manuscript compares field-level inference (FLI) using implicit and explicit methods against power-spectrum-based inference for weak lensing cosmology. Both FLI approaches employ the same 8-million-parameter forward models based on Lagrangian perturbation theory and particle-mesh N-body evolution. The two FLI methods produce closely consistent posteriors, with coverage tests confirming calibration of the implicit analyses. Relative to the power-spectrum analyses, FLI yields significant gains in cosmological information, especially when small scales are included in the PM-based forward model. Remaining challenges for applying explicit PM-based FLI to observational data are discussed.

Significance. If the reported gains hold within the controlled setting, the work is significant for demonstrating a practical, apples-to-apples quantification of map-level information gain over two-point statistics using an identical forward-modeling pipeline. Explicit credit is due for the agreement between implicit and explicit FLI approaches together with the coverage tests that support calibration; these elements strengthen in the field-level constraints. The discussion of remaining challenges for real-data application is also a strength, as it frames the results appropriately for an upcoming-survey context.

minor comments (2)
  1. [Abstract] Abstract: the statement that the results 'show significant gains... especially when small scales are included' would be more informative if it included a quantitative measure (e.g., factor by which credible-interval widths shrink or change in figure-of-merit) rather than a qualitative descriptor.
  2. [Results on small scales] The section on PM forward-model results: the claim that gains increase when small scales are included would benefit from an explicit statement of the PM force-resolution cutoff (in h/Mpc) and a brief note on how this cutoff was chosen relative to the scales where the power-spectrum baseline is evaluated.

Simulated Author's Rebuttal

0 responses · 0 unresolved

We thank the referee for their positive summary of the manuscript, recognition of the significance of the apples-to-apples comparison between field-level and power-spectrum inference, and the recommendation for minor revision. The report correctly notes the consistency between implicit and explicit FLI methods as well as the coverage tests. No specific major comments were provided in the report.

Circularity Check

0 steps flagged

No significant circularity in derivation or comparison chain

full rationale

The paper compares field-level inference against power-spectrum inference on an identical LPT+PM forward model (8M parameters) and reports information gains from the map-level approach. The abstract and described pipeline contain no self-definitional steps, no fitted parameters renamed as predictions, no load-bearing self-citations, and no ansatz smuggling. The central result is a direct empirical comparison of two inference methods applied to the same simulated maps, with internal consistency checks (posterior agreement and coverage tests) that do not reduce to the inputs by construction. This is a standard non-circular methods comparison.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central comparison rests on the assumption that the chosen forward models are adequate representations of reality; no free parameters or invented entities are described in the abstract.

axioms (1)
  • domain assumption Lagrangian perturbation theory and particle-mesh N-body models accurately reproduce the higher-order statistics of weak lensing fields
    The reported gains depend on this fidelity; if the models miss key nonlinear features, the comparison is invalid.

pith-pipeline@v0.9.1-grok · 5781 in / 1080 out tokens · 18145 ms · 2026-06-27T08:47:28.412826+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

104 extracted references · 82 canonical work pages · 11 internal anchors

  1. [1]

    , keywords =

    Data compression and inference in cosmology with self-supervised machine learning. , keywords =. doi:10.1093/mnras/stad3646 , archivePrefix =. 2308.09751 , primaryClass =

    work page doi:10.1093/mnras/stad3646

  2. [2]

    arXiv preprint arXiv:1605.08803 , year=

    Density estimation using real nvp , author=. arXiv preprint arXiv:1605.08803 , year=

    Pith/arXiv arXiv

  3. [3]

    Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

    Deep residual learning for image recognition , author=. Proceedings of the IEEE conference on computer vision and pattern recognition , pages=

  4. [4]

    Fast likelihood-free cosmology with neural density estimators and active learning

    Fast likelihood-free cosmology with neural density estimators and active learning. , keywords =. doi:10.1093/mnras/stz1960 , archivePrefix =. 1903.00007 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stz1960 1903

  5. [5]

    , keywords =

    Beyond the 3rd moment: a practical study of using lensing convergence CDFs for cosmology with DES Y3. , keywords =. doi:10.1093/mnras/stad3118 , archivePrefix =. 2308.03863 , primaryClass =

    work page doi:10.1093/mnras/stad3118

  6. [6]

    , keywords =

    Cosmological constraints using Minkowski functionals from the first year data of the Hyper Suprime-Cam. , keywords =. doi:10.1093/mnras/staf257 , archivePrefix =. 2410.00401 , primaryClass =

    work page doi:10.1093/mnras/staf257

  7. [7]

    , keywords =

    Cosmological cross-correlations and nearest neighbour distributions. , keywords =. doi:10.1093/mnras/stab961 , archivePrefix =. 2102.01184 , primaryClass =

    work page doi:10.1093/mnras/stab961

  8. [8]

    Advances in Neural Information Processing Systems , volume=

    Information maximization in noisy channels: A variational approach , author=. Advances in Neural Information Processing Systems , volume=

  9. [9]

    Weak Gravitational Lensing

    Weak gravitational lensing. , keywords =. doi:10.1016/S0370-1573(00)00082-X , archivePrefix =. astro-ph/9912508 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1016/s0370-1573(00)00082-x

  10. [10]

    arXiv e-prints , keywords =

    Field-Level Inference with Microcanonical Langevin Monte Carlo. arXiv e-prints , keywords =. doi:10.48550/arXiv.2307.09504 , archivePrefix =. 2307.09504 , primaryClass =

    work page doi:10.48550/arxiv.2307.09504

  11. [11]

    Statistics and computing , volume=

    Non-linear regression models for Approximate Bayesian Computation , author=. Statistics and computing , volume=. 2010 , publisher=

    2010

  12. [12]

    , keywords =

    Map-based cosmology inference with weak lensing - information content and its dependence on the parameter space. , keywords =. doi:10.1093/mnrasl/slad160 , archivePrefix =. 2307.00070 , primaryClass =

    work page doi:10.1093/mnrasl/slad160

  13. [13]

    arXiv e-prints , keywords =

    Bayesian mass mapping with weak lensing data using KARMMA -- validation with simulations and application to Dark Energy Survey Year 3 data. arXiv e-prints , keywords =. doi:10.48550/arXiv.2403.05484 , archivePrefix =. 2403.05484 , primaryClass =

    work page doi:10.48550/arxiv.2403.05484

  14. [14]

    James Bradbury and Roy Frostig and Peter Hawkins and others , title =

  15. [15]

    arXiv e-prints , keywords =

    Investigating the Impact of Model Misspecification in Neural Simulation-based Inference. arXiv e-prints , keywords =. doi:10.48550/arXiv.2209.01845 , archivePrefix =. 2209.01845 , primaryClass =

    work page doi:10.48550/arxiv.2209.01845

  16. [16]

    Embedding realistic surveys in simulations through volume remapping

    Embedding Realistic Surveys in Simulations Through Volume Remapping. , keywords =. doi:10.1088/0067-0049/190/2/311 , archivePrefix =. 1003.3178 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0067-0049/190/2/311

  17. [17]

    , keywords =

    Dark Energy Survey Year 1 results: curved-sky weak lensing mass map. , keywords =. doi:10.1093/mnras/stx3363 , adsurl =

    work page doi:10.1093/mnras/stx3363

  18. [18]

    , keywords =

    Weak lensing scattering transform: dark energy and neutrino mass sensitivity. , keywords =. doi:10.1093/mnras/stab2102 , archivePrefix =. 2103.09247 , primaryClass =

    work page doi:10.1093/mnras/stab2102

  19. [19]

    , keywords =

    Cosmological constraints from weak lensing scattering transform using HSC Y1 data. , keywords =. doi:10.1088/1475-7516/2025/01/006 , archivePrefix =. 2404.16085 , primaryClass =

    work page doi:10.1088/1475-7516/2025/01/006 2025

  20. [20]

    , keywords =

    Core Cosmology Library: Precision Cosmological Predictions for LSST. , keywords =. doi:10.3847/1538-4365/ab1658 , archivePrefix =. 1812.05995 , primaryClass =

    work page doi:10.3847/1538-4365/ab1658

  21. [21]

    arXiv e-prints , keywords =

    The LSST Dark Energy Science Collaboration (DESC) Science Requirements Document. arXiv e-prints , keywords =. doi:10.48550/arXiv.1809.01669 , archivePrefix =. 1809.01669 , primaryClass =

    work page doi:10.48550/arxiv.1809.01669

  22. [22]

    Euclid preparation. XXVIII. Forecasts for ten different higher-order weak lensing statistics. , keywords =. doi:10.1051/0004-6361/202346017 , archivePrefix =. 2301.12890 , primaryClass =

    work page doi:10.1051/0004-6361/202346017

  23. [23]

    arXiv preprint arXiv:1506.02169 , year=

    Approximating likelihood ratios with calibrated discriminative classifiers , author=. arXiv preprint arXiv:1506.02169 , year=

    Pith/arXiv arXiv

  24. [24]

    Proceedings of the National Academy of Science , keywords =

    The frontier of simulation-based inference. Proceedings of the National Academy of Science , keywords =. doi:10.1073/pnas.1912789117 , archivePrefix =. 1911.01429 , primaryClass =

    work page doi:10.1073/pnas.1912789117 1911

  25. [25]

    NeurIPS 2024 Workshop: Machine Learning and the Physical Sciences , year=

    Joint cosmological parameter inference and initial condition reconstruction with Stochastic Interpolants , author=. NeurIPS 2024 Workshop: Machine Learning and the Physical Sciences , year=

    2024

  26. [26]

    , keywords =

    Hyper Suprime-Cam Year 3 results: Cosmology from cosmic shear power spectra. , keywords =. doi:10.1103/PhysRevD.108.123519 , archivePrefix =. 2304.00701 , primaryClass =

    work page doi:10.1103/physrevd.108.123519

  27. [27]

    2017 , eprint=

    Density estimation using Real NVP , author=. 2017 , eprint=

    2017

  28. [28]

    , keywords =

    Dark energy survey year 3 results: cosmological constraints from the analysis of cosmic shear in harmonic space. , keywords =. doi:10.1093/mnras/stac1826 , archivePrefix =. 2203.07128 , primaryClass =

    work page doi:10.1093/mnras/stac1826

  29. [29]

    FastPM: a new scheme for fast simulations of dark matter and halos

    FASTPM: a new scheme for fast simulations of dark matter and haloes. , keywords =. doi:10.1093/mnras/stw2123 , archivePrefix =. 1603.00476 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stw2123

  30. [30]

    Cosmological constraints from noisy convergence maps through deep learning , volume=

    Cosmological constraints from noisy convergence maps through deep learning. , keywords =. 2018. doi:10.1103/PhysRevD.98.123518 , primaryClass =

    work page doi:10.1103/physrevd.98.123518 2018

  31. [31]

    astro-ph.IM

    Precision matrix expansion - efficient use of numerical simulations in estimating errors on cosmological parameters. , primaryClass = "astro-ph.IM", keywords =. doi:10.1093/mnras/stx2566 , adsurl =

    work page doi:10.1093/mnras/stx2566

  32. [32]

    , year = 2018, month = jul, volume = 98, number = 2, eid =

    Density split statistics: Joint model of counts and lensing in cells. , year = 2018, month = jul, volume = 98, number = 2, eid =. doi:10.1103/PhysRevD.98.023508 , adsurl =

    work page doi:10.1103/physrevd.98.023508 2018

  33. [33]

    , keywords =

    CFHTLenS: cosmological constraints from a combination of cosmic shear two-point and three-point correlations. , keywords =. 2014. doi:10.1093/mnras/stu754 , primaryClass =

    work page doi:10.1093/mnras/stu754 2014

  34. [34]

    , keywords =

    Dark Energy Survey Year 3 results: cosmology with moments of weak lensing mass maps - validation on simulations. , keywords =. doi:10.1093/mnras/staa2680 , archivePrefix =. 1911.05568 , primaryClass =

    work page doi:10.1093/mnras/staa2680 1911

  35. [35]

    Cosmological results

    Dark Energy Survey Year 3 results: simulation-based cosmological inference with wavelet harmonics, scattering transforms, and moments of weak lensing mass maps II. Cosmological results. arXiv e-prints , keywords =. doi:10.48550/arXiv.2405.10881 , archivePrefix =. 2405.10881 , primaryClass =

    work page doi:10.48550/arxiv.2405.10881

  36. [36]

    2024a, @doi [ ] 10.1103/PhysRevD.109.063534 , https://ui.adsabs.harvard.edu/abs/2024PhRvD.109f3534G 109, 063534

    Dark Energy Survey Year 3 results: Simulation-based cosmological inference with wavelet harmonics, scattering transforms, and moments of weak lensing mass maps. Validation on simulations. , keywords =. doi:10.1103/PhysRevD.109.063534 , archivePrefix =. 2310.17557 , primaryClass =

    work page doi:10.1103/physrevd.109.063534

  37. [37]

    astro-ph.CO

    An improved fitting formula for the dark matter bispectrum. , primaryClass = "astro-ph.CO", year = 2012, month = feb, volume = 2, eid =. doi:10.1088/1475-7516/2012/02/047 , adsurl =

    work page doi:10.1088/1475-7516/2012/02/047 2012

  38. [38]

    , keywords =

    Dark Energy Survey Year 3 Results: Cosmological constraints from second- and third-order shear statistics. , keywords =. doi:10.1103/sxlz-t9gb , archivePrefix =. 2508.14018 , primaryClass =

    work page doi:10.1103/sxlz-t9gb

  39. [39]

    The Open Journal of Astrophysics , keywords =

    Minkowski Functionals in Joint Galaxy Clustering & Weak Lensing Analyses. The Open Journal of Astrophysics , keywords =. doi:10.21105/astro.2206.03877 , archivePrefix =. 2206.03877 , primaryClass =

    work page doi:10.21105/astro.2206.03877

  40. [40]

    , keywords =

    Cosmic shear cosmology beyond two-point statistics: a combined peak count and correlation function analysis of DES-Y1. , keywords =. doi:10.1093/mnras/stab1623 , archivePrefix =. 2012.02777 , primaryClass =

    work page doi:10.1093/mnras/stab1623 2012

  41. [41]

    and Sellentin, Elena and de Mijolla, Damien and Vianello, Alvise , year=

    Massive data compression for parameter-dependent covariance matrices. , keywords =. doi:10.1093/mnras/stx2326 , adsurl =

    work page doi:10.1093/mnras/stx2326

  42. [42]

    doi:10.48550/arXiv.2110.06581 , keywords =

    A Trust Crisis In Simulation-Based Inference? Your Posterior Approximations Can Be Unfaithful. arXiv e-prints , keywords =. doi:10.48550/arXiv.2110.06581 , archivePrefix =. 2110.06581 , primaryClass =

    work page doi:10.48550/arxiv.2110.06581

  43. [43]

    The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo

    The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo. arXiv e-prints , keywords =. doi:10.48550/arXiv.1111.4246 , archivePrefix =. 1111.4246 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1111.4246

  44. [44]

    Hoffman and Andrew Gelman , title =

    Matthew D. Hoffman and Andrew Gelman , title =. Journal of Machine Learning Research , year =

  45. [45]

    , keywords =

    Field-level Neural Network Emulator for Cosmological N-body Simulations. , keywords =. doi:10.3847/1538-4357/acdb6c , archivePrefix =. 2206.04594 , primaryClass =

    work page doi:10.3847/1538-4357/acdb6c

  46. [46]

    Jasche, Jens and Kitaura, Francisco S. , year=. Fast Hamiltonian sampling for large-scale structure inference: Fast Hamiltonian sampling , volume=. Monthly Notices of the Royal Astronomical Society , publisher=. doi:10.1111/j.1365-2966.2010.16897.x , number=

    work page doi:10.1111/j.1365-2966.2010.16897.x 2010

  47. [47]

    Jasche, Jens and Wandelt, Benjamin D. , year=. Bayesian physical reconstruction of initial conditions from large-scale structure surveys , volume=. Monthly Notices of the Royal Astronomical Society , publisher=. doi:10.1093/mnras/stt449 , number=

    work page doi:10.1093/mnras/stt449

  48. [48]

    , keywords =

    Deep learning dark matter map reconstructions from DES SV weak lensing data. , keywords =. doi:10.1093/mnras/staa127 , archivePrefix =. 1908.00543 , primaryClass =

    work page doi:10.1093/mnras/staa127 1908

  49. [49]

    Monthly Notices of the Royal Astronomical Society , volume=

    Likelihood-free inference with neural compression of DES SV weak lensing map statistics , author=. Monthly Notices of the Royal Astronomical Society , volume=. 2021 , publisher=

    2021

  50. [50]

    2026 , publisher =

    Kabalan, Wassim and Lanusse, François and Boucaud, Alexandre and Aubourg, Eric , title =. 2026 , publisher =. doi:10.21105/joss.08852 , url =

    work page doi:10.21105/joss.08852 2026

  51. [51]

    doi:10.1093/mnras/stw2070 , Journal =

    work page doi:10.1093/mnras/stw2070

  52. [52]

    astro-ph.CO

    Information content of weak lensing power spectrum and bispectrum: including the non-Gaussian error covariance matrix. , primaryClass = "astro-ph.CO", keywords =. doi:10.1093/mnras/sts340 , adsurl =

    work page doi:10.1093/mnras/sts340

  53. [53]

    Patrick Kidger , year=

  54. [54]

    Cosmology with cosmic shear observations: a review

    Cosmology with cosmic shear observations: a review. Reports on Progress in Physics , keywords =. doi:10.1088/0034-4885/78/8/086901 , archivePrefix =. 1411.0115 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0034-4885/78/8/086901

  55. [55]

    and Haiman, Zoltán and May, Morgan , year=

    , Keywords =. doi:10.1103/PhysRevD.81.043519 , Eid =

    work page doi:10.1103/physrevd.81.043519

  56. [56]

    , keywords =

    Probing cosmology with weak lensing Minkowski functionals. , keywords =. 2012. doi:10.1103/PhysRevD.85.103513 , primaryClass =

    work page doi:10.1103/physrevd.85.103513 2012

  57. [57]

    , keywords =

    Forecasting the power of higher order weak-lensing statistics with automatically differentiable simulations. , keywords =. doi:10.1051/0004-6361/202346888 , archivePrefix =. 2305.07531 , primaryClass =

    work page doi:10.1051/0004-6361/202346888

  58. [58]

    2025, @doi [ ] 10.1051/0004-6361/202451535 , https://ui.adsabs.harvard.edu/abs/2025A&A...697A.162L 697, A162

    Optimal neural summarization for full-field weak lensing cosmological implicit inference. , keywords =. doi:10.1051/0004-6361/202451535 , archivePrefix =. 2407.10877 , primaryClass =

    work page doi:10.1051/0004-6361/202451535

  59. [59]

    Systematic-free inference of the cosmic matter density field from SDSS3-BOSS data

    Lavaux, Guilhem and Jasche, Jens and Leclercq, Florent. Systematic-free inference of the cosmic matter density field from SDSS3-BOSS data. 2019. arXiv:1909.06396

    arXiv 2019

  60. [60]

    2023 , eprint=

    Sampling-Based Accuracy Testing of Posterior Estimators for General Inference , author=. 2023 , eprint=

    2023

  61. [61]

    arXiv e-prints , keywords =

    pmwd: A Differentiable Cosmological Particle-Mesh N -body Library. arXiv e-prints , keywords =. doi:10.48550/arXiv.2211.09958 , archivePrefix =. 2211.09958 , primaryClass =

    work page doi:10.48550/arxiv.2211.09958

  62. [62]

    Journal of Computational Physics , keywords =

    Perturbation-theory informed integrators for cosmological simulations. Journal of Computational Physics , keywords =. doi:10.1016/j.jcp.2024.113201 , archivePrefix =. 2301.09655 , primaryClass =

    work page doi:10.1016/j.jcp.2024.113201 2024

  63. [63]

    DISCO-DJ II: a differentiable particle-mesh code for cosmology

    DISCO-DJ II: a differentiable particle-mesh code for cosmology. arXiv e-prints , keywords =. doi:10.48550/arXiv.2510.05206 , archivePrefix =. 2510.05206 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2510.05206

  64. [64]

    Cosmological Constraints From Weak Lensing Peak Statistics With CFHT Stripe 82 Survey

    Cosmological constraints from weak lensing peak statistics with Canada-France-Hawaii Telescope Stripe 82 Survey. , keywords =. doi:10.1093/mnras/stv784 , archivePrefix =. 1412.3683 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stv784

  65. [65]

    doi:10.1103/PhysRevD.91.063507 , Eid =

    , Keywords =. doi:10.1103/PhysRevD.91.063507 , Eid =

    work page doi:10.1103/physrevd.91.063507

  66. [66]

    The Open Journal of Astrophysics , keywords =

    Almanac: Weak Lensing power spectra and map inference on the masked sphere. The Open Journal of Astrophysics , keywords =. doi:10.21105/astro.2210.13260 , archivePrefix =. 2210.13260 , primaryClass =

    work page doi:10.21105/astro.2210.13260

  67. [67]

    Weak lensing for precision cosmology

    Weak Lensing for Precision Cosmology. , keywords =. doi:10.1146/annurev-astro-081817-051928 , archivePrefix =. 1710.03235 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1146/annurev-astro-081817-051928

  68. [68]

    , keywords =

    Cosmology from weak lensing peaks and minima with Subaru Hyper Suprime-Cam Survey first-year data. , keywords =. doi:10.1093/mnras/stae098 , archivePrefix =. 2308.10866 , primaryClass =

    work page doi:10.1093/mnras/stae098

  69. [69]

    Constraining cosmology with shear peak statistics: tomographic analysis

    Constraining cosmology with shear peak statistics: tomographic analysis. , keywords =. doi:10.1051/0004-6361/201425164 , archivePrefix =. 1506.02192 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1051/0004-6361/201425164

  70. [70]

    arXiv e-prints , archivePrefix = "arXiv", eprint =

    Dark Energy Survey Year 1 Results: Wide field mass maps via forward fitting in harmonic space. arXiv e-prints , archivePrefix = "arXiv", eprint =

  71. [71]

    Handbook of Markov Chain Monte Carlo , year = 2011, pages =

    MCMC Using Hamiltonian Dynamics. Handbook of Markov Chain Monte Carlo , year = 2011, pages =. doi:10.1201/b10905 , adsurl =

    work page doi:10.1201/b10905 2011

  72. [72]

    , keywords =

    Cosmology from HSC Y1 weak lensing data with combined higher-order statistics and simulation-based inference. , keywords =. doi:10.1103/PhysRevD.111.083510 , archivePrefix =. 2409.01301 , primaryClass =

    work page doi:10.1103/physrevd.111.083510

  73. [73]

    Advances in neural information processing systems , volume=

    Fast -free inference of simulation models with bayesian conditional density estimation , author=. Advances in neural information processing systems , volume=

  74. [74]

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

    Sequential Neural Likelihood: Fast Likelihood-free Inference with Autoregressive Flows , author =. Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics , pages =. 2019 , editor =

    2019

  75. [75]

    , keywords =

    Emulating the CFHTLenS weak lensing data: Cosmological constraints from moments and Minkowski functionals. , keywords =. 2015. doi:10.1103/PhysRevD.91.103511 , primaryClass =

    work page doi:10.1103/physrevd.91.103511 2015

  76. [76]

    Astronomy and Computing , archivePrefix = "arXiv", eprint =

    Mocking the weak lensing universe: The LensTools Python computing package. Astronomy and Computing , archivePrefix = "arXiv", eprint =. doi:10.1016/j.ascom.2016.06.001 , adsurl =

    work page doi:10.1016/j.ascom.2016.06.001 2016

  77. [77]

    , keywords =

    Bayesian forward modelling of cosmic shear data. , keywords =. doi:10.1093/mnras/stab204 , archivePrefix =. 2011.07722 , primaryClass =

    work page doi:10.1093/mnras/stab204 2011

  78. [78]

    , keywords =

    Lifting weak lensing degeneracies with a field-based likelihood. , keywords =. doi:10.1093/mnras/stab3234 , archivePrefix =. 2108.04825 , primaryClass =

    work page doi:10.1093/mnras/stab3234

  79. [79]

    arXiv e-prints , keywords =

    Field-level inference of cosmic shear with intrinsic alignments and baryons. arXiv e-prints , keywords =. doi:10.48550/arXiv.2304.04785 , archivePrefix =. 2304.04785 , primaryClass =

    work page doi:10.48550/arxiv.2304.04785

  80. [80]

    , keywords =

    BULLFROG: multi-step perturbation theory as a time integrator for cosmological simulations. , keywords =. doi:10.1088/1475-7516/2025/02/020 , archivePrefix =. 2409.19049 , primaryClass =

    work page doi:10.1088/1475-7516/2025/02/020 2025

Showing first 80 references.