pith. sign in

arxiv: 2506.03969 · v3 · submitted 2025-06-04 · 🌌 astro-ph.CO

Differentiable Fuzzy Cosmic-Web for Field Level Inference

P. Rossell\'o , F.-S. Kitaura , D. Forero-S\'anchez , F. Sinigaglia , G. Favole This is my paper

Pith reviewed 2026-05-19 10:56 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords field-level inferencecosmic webdifferentiable bias modelBayesian reconstructiongalaxy surveysprimordial density fieldnonlocal biasHICOBIAN
0
0 comments X

The pith

A differentiable fuzzy cosmic-web model reconstructs the primordial density field accurately from galaxy surveys in Bayesian inference.

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

The paper introduces a forward-modeling framework for field-level cosmological inference that couples structure formation with a bias model for galaxy tracers. At its core is the HICOBIAN model, which treats the cosmic web as a fuzzy hierarchy of regions whose transitions are made smooth and differentiable through sigmoid-based gradient operations. This construction permits gradient-based optimization while incorporating long-range and short-range nonlocal bias plus deviations from Poisson statistics. Tests in a self-consistent Bayesian setting show that the method recovers the input primordial density field inside the derived error bars and reconstructs the parameters of an eight-parameter higher-order bias model. The recovered field also saturates nearly the information content allowed by Poisson sampling noise, as confirmed by Fourier-space two- and three-point statistics.

Core claim

By treating transitions between cosmic-web regions as inherently smooth and implementing them via sigmoid-based gradient operations, the HICOBIAN model supplies a differentiable, hierarchical, nonlocal, and stochastic bias description that, when combined with augmented Lagrangian perturbation theory, enables scalable Bayesian field-level inference. This framework recovers the primordial density field within posterior error bars and accurately reconstructs bias parameters for models containing up to eight parameters, while approaching the information limit set by Poisson noise.

What carries the argument

The HICOBIAN model, which supplies a positive-definite tracer field through a fuzzy hierarchical cosmic-web classification that uses sigmoid gradients to remain differentiable while including long- and short-range nonlocal bias plus non-Poisson likelihood terms.

If this is right

  • The primordial density field is recovered inside the error bars derived from full Bayesian posterior sampling.
  • The method extracts nearly the maximum information consistent with Poisson sampling noise.
  • Parameters of a higher-order nonlocal bias model with eight free parameters are reconstructed accurately inside the same Bayesian framework.
  • GPU-accelerated evaluation via the BRIDGE code in JAX makes the model practical for large-volume surveys.

Where Pith is reading between the lines

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

  • The same sigmoid-based fuzzy classification could be ported to other hierarchical partitioning problems in astrophysics that require differentiable forward models.
  • Combining the model with modern machine-learning samplers might further reduce the computational cost of field-level analyses for next-generation surveys.
  • If the smoothness assumption holds across redshift, the framework offers a route to joint inference of initial conditions and bias evolution without ad-hoc region masks.

Load-bearing premise

Transitions between cosmic-web regions are smooth enough that sigmoid approximations introduce negligible bias in the recovered density field and bias parameters.

What would settle it

If the two- and three-point statistics computed from the reconstructed primordial field in Fourier space deviate from the input field by more than the Bayesian posterior error bars, or if the eight bias parameters are not recovered to within their posterior uncertainties, the accuracy claim would be falsified.

Figures

Figures reproduced from arXiv: 2506.03969 by D. Forero-S\'anchez, F. Sinigaglia, F.-S. Kitaura, G. Favole, P. Rossell\'o.

Figure 1
Figure 1. Figure 1: Schematic overview of the BRIDGE pipeline. The framework combines a scalable, differentiable structure formation model with a flexible field-level bias prescription, all implemented in a GPU-accelerated JAX environment. This enables efficient Bayesian inference of the primordial density field from tracer observations, with support for complex, nonlocal bias models and multi-resolution analysis. The modular… view at source ↗
Figure 2
Figure 2. Figure 2: Example of fuzzy cosmic-web classification with N = 256 and ∆L = 1.7 h −1Mpc. Top left: evolved dark matter density contrast field. Bottom left: “hard” Φ-web classification. Right panels: fuzzy membership weights p (V) i , p (S) i , p (F) i , and p (K) i for voids, sheets, filaments, and knots, respectively, as defined in Eq. 16. 2.8. Differentiable Fuzzy Cosmic-Web Classification In the top-down structure… view at source ↗
Figure 3
Figure 3. Figure 3: TEST1: Mean autocorrelation ξ(n) as a function of lag n for a chain of length M ≈ 1000, evaluated over 104 randomly selected vox￾els. The red curve shows the ensemble average ⟨ξ(n)⟩ across parame￾ters, with the shaded band indicating the ±1σ dispersion. The horizontal dashed line marks the threshold ξ = 0.1 used to assess mixing efficiency. 32 64 128 256 N 10 3 10 2 10 1 Gradient Time (s) CWEB None web web… view at source ↗
Figure 4
Figure 4. Figure 4: Computing times for single gradient evaluations of the for￾ward model employing ALPT evolution and the HICOBIAN bias model for different mesh sizes, run on a single NVIDIA A100-SXM4 GPU equipped with 40 GB of on-board HBM2 memory. 10 2 10 1 k [Mpc h 1 ] 1.00 1.05 1.10 1.15 R ( 0 ) knyq 50% R( 0) = 1.013 75% 90% 99% [PITH_FULL_IMAGE:figures/full_fig_p009_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Gelman-Rubin convergence diagnostic computed for three in￾dependently initialized chains of 250 samples after convergence. This statistic is calculated for all cells in Fourier space, and we show their distribution with scale k. Values close to one indicate that the Markov chains have converged and that the samples are effectively independent. deviations of the reconstructed fields display structured patte… view at source ↗
Figure 6
Figure 6. Figure 6: TEST1: Spatial summary of 500 independent reconstructions (N = 128). Slices of ∆L = 10 h −1Mpc (one voxel width). Top row: recon￾structed initial density contrast. Bottom row: reconstructed tracer field. Columns from left to right: (1) the pixel-wise standard deviation across the 500 independent reconstructions, (2) the mean reconstructed field, (3) one representative reconstruction sample, and (4) the tru… view at source ↗
Figure 7
Figure 7. Figure 7: TEST1: Comparison of the summary statistics of the initial density contrast field for the mock reference (red dashed) and 500 independent reconstructions (blue solid). Columns from left to right: (1) the monopole power spectrum, (2) the quadrupole, (3) the reduced bispectrum as a function of triangle opening angle θ, and (4) the cross-correlation coefficient C(k) between the reference and reconstruction (b… view at source ↗
Figure 8
Figure 8. Figure 8: TEST1: Comparison of the summary statistics of the mock reference, and 500 independent reconstructions evolved through the forward model. Panel definitions are identical to those in [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: TEST1: Posterior distributions for the 500 independent samples of the α and β bias parameters in each cosmic-web region of the Φ-web. Red dots and lines show the values of the parameters for the ground-truth catalogue. The posteriors are smoothed with a 2D Gaussian filter with standard deviation equal to 0.75 bin widths. TEST2 yields equivalent results. plex bias scenarios involving 4 to 16 cosmic-web regi… view at source ↗
read the original abstract

A comprehensive analysis of the cosmological large-scale structure derived from galaxy surveys involves field-level inference, which requires a forward modelling framework that simultaneously accounts for structure formation and tracer bias. While structure formation models are well-understood, the development of an effective field-level bias model remains challenging within Bayesian reconstruction methods, which we address in this work. To bridge this gap, we have developed a differentiable model that integrates augmented Lagrangian perturbation theory, nonlinear, nonlocal, and stochastic biasing. At the core of our approach is the HICOBIAN model, which provides a description of a field with a positive number of tracers while incorporating a long- and short-range nonlocal framework and deviations from Poissonity in the likelihood. A key insight of our model is that transitions between cosmic-web regions are inherently smooth, which we implement using sigmoid-based gradient operations. This enables a fuzzy, and, hence, differentiable hierarchical cosmic-web description, making the model well-suited for machine learning frameworks. We test the practical implementation of this model through GPU-accelerated computations implemented in JaX, the BRIDGE code, enabling scalable evaluation of complex biasing. Our approach accurately reproduces the primordial density field within associated error bars derived from Bayesian posterior sampling within a self-specified setting as validated by two- and three-point statistics in Fourier space. Furthermore, we demonstrate that the methodology approaches the maximum encoded information consistent with Poisson noise. We also demonstrate that the bias parameters of a higher-order nonlocal bias model can be accurately reconstructed within the Bayesian framework for bias models with eight parameters.

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 / 2 minor

Summary. The paper introduces a differentiable forward model for cosmological field-level inference that combines augmented Lagrangian perturbation theory (ALPT) with the HICOBIAN bias model. The HICOBIAN model incorporates long- and short-range nonlocal biasing, stochasticity, and deviations from Poisson statistics for positive tracer counts. Smooth transitions between cosmic-web regions are enforced via sigmoid-based gradient operations to produce a fuzzy, hierarchical, and fully differentiable description. The model is implemented in JAX (BRIDGE code) for GPU acceleration. Validation on self-consistent mocks generated under the same ALPT + HICOBIAN assumptions shows recovery of the primordial density field within Bayesian posterior error bars, agreement with two- and three-point Fourier-space statistics, approach to the Poisson-noise information limit, and accurate reconstruction of up to eight higher-order nonlocal bias parameters.

Significance. If the central claims hold, the work provides a scalable, gradient-friendly framework that integrates structure formation and bias modeling for field-level Bayesian inference. The JAX implementation and emphasis on differentiability are strengths that could facilitate integration with machine-learning pipelines. The self-consistent recovery tests and demonstration of approaching the Poisson limit are positive indicators of internal performance, though the significance is tempered by the lack of tests against model misspecification.

major comments (2)
  1. [Validation / Results] Validation section (as described in the abstract and results): The recovery of the primordial density field within posterior error bars and the reconstruction of eight bias parameters are demonstrated exclusively on mocks generated with the identical ALPT + HICOBIAN model and the same sigmoid smoothing. This establishes internal consistency but does not test robustness when the true tracer field contains sharper cosmic-web transitions or a different higher-order bias form. Because the central claim is that the method accurately reproduces the field 'within associated error bars' in a self-specified setting, this limitation is load-bearing; an additional test suite with misspecified bias models is required to support the claim that the posterior errors are reliable.
  2. [Methods / HICOBIAN model] Methods section on the HICOBIAN model and sigmoid operations: The premise that cosmic-web transitions are inherently smooth is implemented via sigmoid-based gradients to enable differentiability. No quantitative assessment is provided of the bias introduced by this approximation relative to a sharp classification when the underlying density field contains steeper gradients. If this approximation systematically shifts the effective likelihood, the MAP or posterior mean for the initial conditions could lie outside the reported error bars even while low-order Fourier statistics remain matched.
minor comments (2)
  1. [Abstract / Introduction] The abstract and introduction would benefit from a clearer statement of the precise form of the Poisson-like likelihood and how deviations from Poissonity are parameterized within HICOBIAN.
  2. [Figures] Figure captions for the Fourier-space validation plots should explicitly state the number of realizations, the k-range used for the two- and three-point statistics, and whether the error bars include cosmic variance or only shot noise.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed report. We address each major comment below, clarifying the scope of our claims and indicating where revisions will be made.

read point-by-point responses

  1. Referee: [Validation / Results] Validation section (as described in the abstract and results): The recovery of the primordial density field within posterior error bars and the reconstruction of eight bias parameters are demonstrated exclusively on mocks generated with the identical ALPT + HICOBIAN model and the same sigmoid smoothing. This establishes internal consistency but does not test robustness when the true tracer field contains sharper cosmic-web transitions or a different higher-order bias form. Because the central claim is that the method accurately reproduces the field 'within associated error bars' in a self-specified setting, this limitation is load-bearing; an additional test suite with misspecified bias models is required to support the claim that the posterior errors are reliable.

    Authors: The manuscript explicitly limits its validation claims to a 'self-specified setting' (abstract and results section), where the forward model matches the data generation process exactly. This is a standard and necessary first step to confirm that the differentiable Bayesian framework recovers the input fields and parameters without artifacts from the implementation or sampling. We agree that robustness under model misspecification would provide stronger evidence for the reliability of the reported posterior uncertainties in more general cases. We will revise the discussion to more explicitly highlight this scope and note that misspecification tests are an important direction for future work. revision: partial

  2. Referee: [Methods / HICOBIAN model] Methods section on the HICOBIAN model and sigmoid operations: The premise that cosmic-web transitions are inherently smooth is implemented via sigmoid-based gradients to enable differentiability. No quantitative assessment is provided of the bias introduced by this approximation relative to a sharp classification when the underlying density field contains steeper gradients. If this approximation systematically shifts the effective likelihood, the MAP or posterior mean for the initial conditions could lie outside the reported error bars even while low-order Fourier statistics remain matched.

    Authors: The sigmoid smoothing is introduced to reflect the physical expectation that cosmic-web region transitions occur over a finite density range rather than at infinitely sharp thresholds, while simultaneously ensuring full differentiability. The steepness parameter is selected to be consistent with the gradient scales present in the ALPT density fields used throughout the analysis. Although the current manuscript does not include a direct side-by-side quantification against a non-differentiable sharp classifier, the close agreement of the reconstructed fields with two- and three-point Fourier statistics, together with unbiased recovery of the eight bias parameters, indicates that any residual approximation error remains within the posterior uncertainties. We will add a concise justification of the sigmoid choice and its controlled impact in the revised methods section. revision: partial

Circularity Check

0 steps flagged

No significant circularity; internal consistency test is standard and non-reductive

full rationale

The paper constructs a differentiable forward model (ALPT + HICOBIAN with sigmoid-based fuzzy cosmic-web transitions) explicitly to enable gradient-based Bayesian inference. The central claims concern recovery of the initial density field and up to eight bias parameters from mocks generated under identical model assumptions, with validation via Fourier-space 2- and 3-point statistics and proximity to the Poisson-noise information limit. This is a conventional self-consistency check for an inference pipeline rather than a derivation that reduces to its inputs by construction. No equations are shown to equate a claimed prediction with a fitted quantity, no load-bearing self-citation chain is invoked to justify uniqueness, and the sigmoid smoothness is presented as an explicit modeling choice for differentiability, not derived from or equivalent to the target result. The framework remains self-contained against its stated benchmarks.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 1 invented entities

The model rests on several modeling choices and new constructs whose independent support is not provided outside the paper itself. The fuzzy description and specific form of the likelihood are introduced to achieve differentiability and positive tracer counts.

free parameters (1)
  • higher-order nonlocal bias parameters = eight parameters
    Eight parameters of the higher-order nonlocal bias model are reconstructed within the Bayesian framework; these are fitted quantities whose values are not derived from first principles.
axioms (2)
  • domain assumption Transitions between cosmic-web regions are inherently smooth
    Invoked to justify the use of sigmoid-based gradient operations for a differentiable hierarchical description.
  • ad hoc to paper The HICOBIAN model provides a sufficient description of fields with positive tracer counts incorporating long- and short-range nonlocal effects and deviations from Poissonity
    Core modeling assumption stated as the basis for the likelihood and bias framework.
invented entities (1)
  • HICOBIAN model no independent evidence
    purpose: Provides a differentiable description of a field with positive number of tracers using long- and short-range nonlocal framework and deviations from Poissonity in the likelihood
    New model introduced to bridge structure formation and tracer bias in field-level inference.

pith-pipeline@v0.9.0 · 5821 in / 1853 out tokens · 36952 ms · 2026-05-19T10:56:18.253257+00:00 · methodology

discussion (0)

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

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

99 extracted references · 99 canonical work pages · 5 internal anchors

  1. [1]

    2016 [arXiv:1606.00180]

    Amendola, L., Appleby, S., Avgoustidis, A., & others. 2016 [arXiv:1606.00180]

    work page arXiv 2016

  2. [2]

    2023, MNRAS, 520, 5746

    Andrews, A., Jasche, J., Lavaux, G., & Schmidt, F. 2023, MNRAS, 520, 5746

    work page 2023

  3. [3]

    Angulo, R. E. & Hahn, O. 2022, LRCA, 8, 1

    work page 2022

  4. [4]

    2015, MNRAS, 446, 4250 Balaguera-Antolínez, A., Kitaura, F.-S., Alam, S., et al

    Ata, M., Kitaura, F.-S., & Müller, V . 2015, MNRAS, 446, 4250 Balaguera-Antolínez, A., Kitaura, F.-S., Alam, S., et al. 2023, A&A, 673, A130 Balaguera-Antolínez, A., Kitaura, F.-S., Pellejero-Ibáñez, M., et al. 2020, MN- RAS, 491, 2565 Balaguera-Antolínez, A., Kitaura, F.-S., Pellejero-Ibáñez, M., Zhao, C., & Abel, T. 2018, MNRAS: Letters, 483, L58

    work page 2015

  5. [5]

    E., Hamaus, N., & Desjacques, V

    Baldauf, T., Seljak, U., Smith, R. E., Hamaus, N., & Desjacques, V . 2013, Phys. Rev. D, 88, 083507

    work page 2013

  6. [6]

    1994, ApJ, 427, 51

    Bernardeau, F. 1994, ApJ, 427, 51

    work page 1994

  7. [7]

    2002, Physics Reports, 367, 1

    Bernardeau, F., Colombi, S., Gaztañaga, E., & Scoccimarro, R. 2002, Physics Reports, 367, 1

    work page 2002

  8. [8]

    & Dekel, A

    Bertschinger, E. & Dekel, A. 1989, ApJ, 336, L5

    work page 1989

  9. [9]

    & Davidsen, A

    Bi, H. & Davidsen, A. F. 1997, ApJ, 479, 523

    work page 1997

  10. [10]

    2019, MNRAS, 485, 2806

    Blot, L., Crocce, M., Sefusatti, E., et al. 2019, MNRAS, 485, 2806

    work page 2019

  11. [11]

    R., Kofman, L., & Pogosyan, D

    Bond, J. R., Kofman, L., & Pogosyan, D. 1996, Nature, 380, 603

    work page 1996

  12. [12]

    Bos, E. G. P., Kitaura, F.-S., & van de Weygaert, R. 2019, MNRAS, 488, 2573

    work page 2019

  13. [13]

    R., Colombi, S., Hivon, E., & Juszkiewicz, R

    Bouchet, F. R., Colombi, S., Hivon, E., & Juszkiewicz, R. 1995, A&A, 296, 575

    work page 1995

  14. [14]

    2018, http://github.com/jax-ml/jax

    Bradbury, J., Frostig, R., Hawkins, P., et al. 2018, http://github.com/jax-ml/jax

    work page 2018

  15. [15]

    1994, MNRAS, 267, 811

    Buchert, T. 1994, MNRAS, 267, 811

    work page 1994

  16. [16]

    J., Sheth, R

    Casas-Miranda, R., Mo, H. J., Sheth, R. K., & Boerner, G. 2002, MNRAS, 333, 730–738

    work page 2002

  17. [17]

    1995, MNRAS, 276, 115

    Catelan, P. 1995, MNRAS, 276, 115

    work page 1995

  18. [18]

    & Ostriker, J

    Cen, R. & Ostriker, J. P. 1993, ApJ, 417, 415

    work page 1993

  19. [19]

    C., Scoccimarro, R., & Sheth, R

    Chan, K. C., Scoccimarro, R., & Sheth, R. K. 2012, Phys. Rev. D, 85, 083509

    work page 2012

  20. [20]

    2017, Phys

    Chuang, C.-H., Kitaura, F.-S., Liang, Y ., et al. 2017, Phys. Rev. D, 95, 063528

    work page 2017

  21. [21]

    2019, MNRAS, 487, 48

    Chuang, C.-H., Yepes, G., Kitaura, F.-S., et al. 2019, MNRAS, 487, 48

    work page 2019

  22. [22]

    & Jones, B

    Coles, P. & Jones, B. 1991, MNRAS, 248, 1

    work page 1991

  23. [23]

    M., Kitaura, F.-S., García-Farieta, J

    Coloma-Nadal, J. M., Kitaura, F.-S., García-Farieta, J. E., et al. 2024, JCAP, 2024, 083 Conceição, M., Krone-Martins, A., da Silva, A., & Moliné, Á. 2024, A&A, 681, A123

    work page 2024

  24. [24]

    Data Release 1 of the Dark Energy Spectroscopic Instrument

    Dekel, A. & Lahav, O. 1999, ApJ, 520, 24 DESI Collaboration, Abdul-Karim, M., Adame, A. G., et al. 2025, arXiv:2503.14745

    work page internal anchor Pith review Pith/arXiv arXiv 1999

  25. [25]

    2018, Phys

    Desjacques, V ., Jeong, D., & Schmidt, F. 2018, Phys. Rep., 733, 1

    work page 2018

  26. [26]

    2024, 535, 1258

    Doeser, L., Jamieson, D., Stopyra, S., et al. 2024, 535, 1258

    work page 2024

  27. [27]

    J., & Roweth, D

    Duane, S., Kennedy, A., Pendleton, B. J., & Roweth, D. 1987, Phys. Letters B, 195, 216

    work page 1987

  28. [28]

    J., Seo, H.-J., Sirko, E., & Spergel, D

    Eisenstein, D. J., Seo, H.-J., Sirko, E., & Spergel, D. N. 2007, ApJ, 664, 675

    work page 2007

  29. [29]

    Favole, G. et al. 2025, in preparation

    work page 2025

  30. [30]

    2016, MNRAS, 463, 2273

    Feng, Y ., Chu, M.-Y ., Seljak, U., & McDonald, P. 2016, MNRAS, 463, 2273

    work page 2016

  31. [31]

    E., Sánchez Almeida, J., Muñoz-Tuñón, C., et al

    Filho, M. E., Sánchez Almeida, J., Muñoz-Tuñón, C., et al. 2015, ApJ, 802, 82 Forero Sánchez, D., Kitaura, F. S., Sinigaglia, F., Coloma-Nadal, J. M., & Kneib, J. P. 2024, J. Cosmology Astropart. Phys., 2024, 001 Forero–Romero, J. E., Hoffman, Y ., Gottlöber, S., Klypin, A., & Yepes, G. 2009, MNRAS, 396, 1815

    work page 2015

  32. [32]

    Fry, J. N. & Gaztanaga, E. 1993, ApJ, 413, 447 García-Farieta, J. E., Balaguera-Antolínez, A., & Kitaura, F.-S. 2024, A&A, 690, A27

    work page 1993

  33. [33]

    H., Eisenstein, D

    Garrison, L. H., Eisenstein, D. J., Ferrer, D., et al. 2018, ApJS, 236, 43

    work page 2018

  34. [34]

    & Villaescusa-Navarro, F

    Hahn, C. & Villaescusa-Navarro, F. 2021, JCAP, 2021, 029

    work page 2021

  35. [35]

    M., & Dekel, A

    Hahn, O., Porciani, C., Carollo, C. M., & Dekel, A. 2007, MNRAS, 375, 489

    work page 2007

  36. [36]

    Hamilton, A. J. S. 1998, in Astrophysics and Space Science Library, V ol. 231, The Evolving Universe, ed. D. Hamilton, 185

    work page 1998

  37. [37]

    & Peacock, J

    Heavens, A. & Peacock, J. 1988, MNRAS, 232, 339 Heß, S., Kitaura, F.-S., & Gottlöber, S. 2013, MNRAS, 435, 2065

    work page 1988

  38. [38]

    Hoffman, M. D. & Gelman, A. 2011, arXiv:1111.4246

    work page internal anchor Pith review Pith/arXiv arXiv 2011

  39. [39]

    2022, ApJ, 941, 42

    Horowitz, B., Dornfest, M., Luki´c, Z., & Harrington, P. 2022, ApJ, 941, 42

    work page 2022

  40. [40]

    & Lukic, Z

    Horowitz, B. & Lukic, Z. 2025 [2502.02294]

    work page arXiv 2025

  41. [41]

    2019b, JCAP, 2019, 035

    Horowitz, B., Seljak, U., & Aslanyan, G. 2019b, JCAP, 2019, 035

    work page 2019

  42. [42]

    2021, ApJ, 906, 110

    Horowitz, B., Zhang, B., Lee, K.-G., & Kooistra, R. 2021, ApJ, 906, 110

    work page 2021

  43. [43]

    & Gnedin, N

    Hui, L. & Gnedin, N. Y . 1997, MNRAS, 292, 27

    work page 1997

  44. [44]

    A., et al

    Ishiyama, T., Prada, F., Klypin, A. A., et al. 2021, MNRAS, 506, 4210

    work page 2021

  45. [45]

    M., Obuljen, A., Cuesta-Lazaro, C., & Toomey, M

    Ivanov, M. M., Obuljen, A., Cuesta-Lazaro, C., & Toomey, M. W. 2025, Phys. Rev. D, 111, 063548

    work page 2025

  46. [46]

    & Kitaura, F.-S

    Jasche, J. & Kitaura, F.-S. 2010, MNRAS, 407, 29

    work page 2010

  47. [47]

    S., Li, C., & Enßlin, T

    Jasche, J., Kitaura, F. S., Li, C., & Enßlin, T. A. 2010, MNRAS, 409, 355

    work page 2010

  48. [48]

    & Lavaux, G

    Jasche, J. & Lavaux, G. 2019, A&A, 625, A64

    work page 2019

  49. [49]

    2015, 2015, 036

    Jasche, J., Leclercq, F., & Wandelt, B. 2015, 2015, 036

    work page 2015

  50. [50]

    & Wandelt, B

    Jasche, J. & Wandelt, B. D. 2013, MNRAS, 432, 894

    work page 2013

  51. [51]

    1987, MNRAS, 227, 1

    Kaiser, N. 1987, MNRAS, 227, 1

    work page 1987

  52. [52]

    2013, MNRAS: Letters, 429, L84

    Kitaura, F.-S. 2013, MNRAS: Letters, 429, L84

    work page 2013

  53. [53]

    A., et al

    Kitaura, F.-S., Ata, M., Rodríguez-Torres, S. A., et al. 2021, MNRAS, 502, 3456

    work page 2021

  54. [54]

    2022, MNRAS, 512, 2245

    Kitaura, F.-S., Balaguera-Antolínez, A., Sinigaglia, F., & Pellejero-Ibáñez, M. 2022, MNRAS, 512, 2245

    work page 2022

  55. [55]

    & Enßlin, T

    Kitaura, F.-S. & Enßlin, T. A. 2008, MNRAS, 389, 497

    work page 2008

  56. [56]

    G., et al

    Kitaura, F.-S., Gil-Marín, H., Scóccola, C. G., et al. 2015, MNRAS, 450, 1836

    work page 2015

  57. [57]

    & Heß, S

    Kitaura, F.-S. & Heß, S. 2013, MNRAS: Letters, 435, L78

    work page 2013

  58. [58]

    Kitaura, F.-S., Jasche, J., & Metcalf, R. B. 2010, MNRAS, 403, 589

    work page 2010

  59. [59]

    2024, A&A, 683, A215

    Kitaura, F.-S., Sinigaglia, F., Balaguera-Antolínez, A., & Favole, G. 2024, A&A, 683, A215

    work page 2024

  60. [60]

    2014, MNRAS: Letters, 439, L21

    Kitaura, F.-S., Yepes, G., & Prada, F. 2014, MNRAS: Letters, 439, L21

    work page 2014

  61. [61]

    & Prada, F

    Klypin, A. & Prada, F. 2018, MNRAS, 478, 4602 Kodi Ramanah, D., Charnock, T., Villaescusa-Navarro, F., & Wandelt, B. D. 2020, MNRAS, 495, 4227

    work page 2018

  62. [62]

    & Jasche, J

    Lavaux, G. & Jasche, J. 2016, 455, 3169

    work page 2016

  63. [63]

    2019, Systematic-Free Inference of the Cosmic Matter Density Field from SDSS3-BOSS Data

    Lavaux, G., Jasche, J., & Leclercq, F. 2019, Systematic-Free Inference of the Cosmic Matter Density Field from SDSS3-BOSS Data

    work page 2019

  64. [64]

    The DESI Experiment, a whitepaper for Snowmass 2013

    Levi, M., Bebek, C., Beers, T., et al. 2013 [arXiv:1308.0847]

    work page internal anchor Pith review Pith/arXiv arXiv 2013

  65. [65]

    2024, ApJS, 270, 36

    Li, Y ., Modi, C., Jamieson, D., et al. 2024, ApJS, 270, 36

    work page 2024

  66. [66]

    2000, ApJ, 543, 1

    McDonald, P., Miralda-Escudé, J., Rauch, M., et al. 2000, ApJ, 543, 1

    work page 2000

  67. [67]

    & Roy, A

    McDonald, P. & Roy, A. 2009, JCAP, 2009, 020

    work page 2009

  68. [68]

    2021, Astronomy and Computing, 37, 100505

    Modi, C., Lanusse, F., & Seljak, U. 2021, Astronomy and Computing, 37, 100505

    work page 2021

  69. [69]

    & Efstathiou, G

    Monaco, P. & Efstathiou, G. 1999, MNRAS, 308, 763

    work page 1999

  70. [70]

    Neyrinck, M. C. 2016, MNRAS, 455, L11

    work page 2016

  71. [71]

    C., Aragón-Calvo, M

    Neyrinck, M. C., Aragón-Calvo, M. A., Jeong, D., & Wang, X. 2014, MNRAS, 441, 646

    work page 2014

  72. [72]

    & Dekel, A

    Nusser, A. & Dekel, A. 1992, ApJ, 391, 443

    work page 1992

  73. [73]

    E., Kitaura, F.-S., Heß, S., Libeskind, N

    Nuza, S. E., Kitaura, F.-S., Heß, S., Libeskind, N. I., & Müller, V . 2014, MNRAS, 445, 988

    work page 2014

  74. [74]

    A., & Knebe, A

    Olex, E., Hellwing, W. A., & Knebe, A. 2025, 696, A142

    work page 2025

  75. [75]

    2025, JCAP, 2025, 142

    Paillas, E., Ding, Z., Chen, X., et al. 2025, JCAP, 2025, 142

    work page 2025

  76. [76]

    Peebles, P. J. E. 1980, The large-scale structure of the universe (Princeton Uni- versity Press) Pellejero-Ibañez, M., Balaguera-Antolínez, A., Kitaura, F.-S., et al. 2020, MN- RAS, 493, 586

    work page 1980

  77. [77]

    Composable Effects for Flexible and Accelerated Probabilistic Programming in NumPyro

    Phan, D., Pradhan, N., & Jankowiak, M. 2019 [1912.11554]

    work page internal anchor Pith review Pith/arXiv arXiv 2019

  78. [78]

    2019, 630, A151 Rodríguez-Torres, S

    Porqueres, N., Jasche, J., Lavaux, G., & Enßlin, T. 2019, 630, A151 Rodríguez-Torres, S. A., Chuang, C.-H., Prada, F., et al. 2016, MNRAS, 460, 1173

    work page 2019

  79. [79]

    Saslaw, W. C. & Hamilton, A. J. S. 1984, ApJ, 276, 13

    work page 1984

  80. [80]

    2019, Phys

    Schmittfull, M., Simonovi ´c, M., Assassi, V ., & Zaldarriaga, M. 2019, Phys. Rev. D, 100, 043514

    work page 2019

Showing first 80 references.