pith. sign in

arxiv: 2606.12938 · v1 · pith:5AE65KU6new · submitted 2026-06-11 · 🌌 astro-ph.CO

Cluster Mass Inference from Galaxy Kinematics

Bonny Y. Wang , Leander Thiele , Matthew Ho This is my paper

Pith reviewed 2026-06-27 06:14 UTC · model grok-4.3

classification 🌌 astro-ph.CO
keywords galaxy clustersmass estimationphase-spacesimulation-based inferenceneural networksvelocity dispersioninterloperscosmology
0
0 comments X

The pith

A neural network trained on simulations infers cluster masses from galaxy positions and velocities with half the usual scatter.

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

The paper builds a pipeline that feeds the full projected positions and velocities of galaxies into a set-based neural model to predict corrections on top of the classical velocity dispersion mass relation. By training on simulated clusters it learns to extract additional information beyond dispersion alone and outputs full posterior distributions. In clean idealized data the scatter falls to roughly 0.1 dex; the same model keeps comparable accuracy at the high-mass end even when realistic cylindrical projections include interlopers. The approach therefore saturates the kinematic information available from galaxy phase space.

Core claim

The central claim is that a permutation-invariant Deep Sets architecture combined with neural posterior estimation via normalizing flows recovers cluster masses by learning explicit residual corrections to the classical M-σ relation, reducing scatter to approximately 0.1 dex in idealized interloper-free cases and maintaining performance in realistic cylindrical observations at masses above 10^14.5 solar masses per h.

What carries the argument

Permutation-invariant Deep Sets architecture with normalizing-flow posterior estimation that isolates kinematic information beyond velocity dispersion.

If this is right

  • Mass estimates improve enough to tighten cosmological constraints from upcoming cluster surveys.
  • Full posterior outputs supply reliable uncertainty quantification for each cluster.
  • Performance remains stable against interloper contamination in high-mass systems.
  • The kinematic information content is saturated, establishing a baseline for later refinements.

Where Pith is reading between the lines

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

  • The same set-based architecture could be retrained on other cluster observables such as richness or X-ray temperature to extract complementary information.
  • Cross-validation against independent mock catalogs would test whether the learned corrections transfer beyond the training simulation.
  • Deployment on wide-field surveys would allow direct tests of whether the reduced scatter improves dark-energy constraints from cluster abundance.

Load-bearing premise

The simulation used for training accurately captures how real galaxy positions and velocities relate to true cluster mass.

What would settle it

Applying the trained model to real observed clusters and checking whether the inferred masses match independent weak-lensing measurements to within the predicted 0.1 dex scatter.

Figures

Figures reproduced from arXiv: 2606.12938 by Bonny Y. Wang, Leander Thiele, Matthew Ho.

Figure 1
Figure 1. Figure 1: — Posterior probability distributions for galaxy cluster masses. Solid blue curves represent the outputs of the neural pos￾terior estimator (NPE) from this work, while dashed blue lines indicate Gaussian fits to the inferred posteriors. The true cluster masses are shown as red vertical lines, and the dotted orange lines denote estimates based on the classical M–σ relation. The left column displays represen… view at source ↗
Figure 3
Figure 3. Figure 3: — Relationship between the virial mass Mvir and the random line-of-sight velocity dispersion σv of member galaxies in our dataset (idealized setup). The color scale represents the number density of clusters in each bin. The data approximately follow a linear trend, and the best-fit regression line is given by log10(σv) = 0.33 log10(Mvir) − 1.95 [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: — Comparison of cluster mass functions. The black dashed line shows the original distribution from the simulation, while the red solid line represents the augmented data, which are uniformly distributed. To assess how much additional information can be ex￾tracted from galaxy dynamics beyond the baseline M–σ relationship, we train a machine learning model to pre￾dict the posterior density over the residual … view at source ↗
Figure 5
Figure 5. Figure 5: — Schematic overview of the model for interloper identification and probabilistic cluster mass estimation. Galaxy observables— line-of-sight velocity vLOS, projected radius Rradial, and angular position α—are constructed by projecting galaxies onto randomly oriented planes for both idealized and cylindrical catalogs. For mass inference (red arrows), these features are processed through three Deep Sets bloc… view at source ↗
Figure 6
Figure 6. Figure 6: — Comparison of predicted and true galaxy cluster masses. The x-axis represents the true mass values and the y-axis shows the predicted masses obtained through our machine learning. The orange points mark the viral prediction from galaxies’ velocity dispersion, while the violin plots illustrate the generated posterior distributions. The black reference line shows when true and predicted values perfectly ma… view at source ↗
Figure 7
Figure 7. Figure 7: — Tests of the trained models’ performance. Left: Comparison of coverage probabilities against credibility levels for TARP and HPD methods, where HPD (orange) represents the highest posterior density coverage test and TARP (green) represents the distance to random point method. Right: Bias and scatter in predicted cluster masses as a function of true mass. The orange and blue curves represent the standard … view at source ↗
Figure 8
Figure 8. Figure 8: — Predicted posterior standard deviation as a function of true cluster mass, color-coded by normalized cluster concentration. The overlaid contours indicate the 68%, 90%, 94%, and 98% levels of the number density distribution, revealing a bimodal structure: the upper peak corresponds to higher concentrations, while the lower peak corresponds to lower concentrations. may explain how the bimodal structure be… view at source ↗
Figure 9
Figure 9. Figure 9: — Three-dimensional phase-space distribution of galaxies within a cylindrical aperture around a representative cluster. Blue points indicate true cluster members identified by the Rockstar halo finder, while red points denote interloper galaxies. All galax￾ies also satisfy log10(M∗/h−1M⊙) ≥ 9.5. The axes show the calcu￾lated velocity ∆v and the two projected spatial coordinates xproj and yproj. The green s… view at source ↗
read the original abstract

The masses of galaxy clusters carry cosmological and astrophysical information. We develop a simulation-based inference pipeline to infer cluster masses from full projected phase-space information of member and interloper galaxies. Our method combines a permutation-invariant Deep Sets architecture with neural posterior estimation using normalizing flows, enabling the recovery of expressive posterior distributions. We train the model to predict residual corrections to the classical $M$--$\sigma$ relation, thus explicitly isolating information beyond velocity dispersion. Using the Uchuu-UniverseMachine simulation, we evaluate the method under both idealized (interloper-free) and realistic (cylindrical) observational setups. In the idealized case, our model reduces the scatter in mass estimates to as low as $\sim 0.1$ dex, representing a twofold improvement over the traditional $M$--$\sigma$ relation. In the cylindrical setup, we achieve comparable performance at the high-mass end ($> 10^{14.5}\,M_\odot/h$), demonstrating robustness against interloper contamination. We demonstrate that set-based simulation-driven inference provides a powerful and flexible framework for galaxy cluster mass estimation, enabling improved accuracy and reliable uncertainty characterization for upcoming large-scale surveys. Our model saturates the kinematic information content and thus suggests a baseline for future studies.

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

1 major / 1 minor

Summary. The manuscript presents a simulation-based inference pipeline combining a permutation-invariant Deep Sets architecture with neural posterior estimation via normalizing flows to infer galaxy cluster masses from projected phase-space data of member and interloper galaxies. The model is trained on the Uchuu-UniverseMachine simulation to predict residual corrections to the classical M–σ relation, with reported scatter reductions to ~0.1 dex (twofold improvement) in idealized interloper-free cases and comparable high-mass performance (>10^14.5 M_⊙/h) in realistic cylindrical setups.

Significance. If the simulation-based corrections generalize, the approach offers a flexible framework for more accurate mass estimates with reliable uncertainty quantification, leveraging full phase-space information beyond velocity dispersion alone. The explicit residual modeling and permutation-invariant architecture are methodological strengths that could serve as a baseline for future survey analyses.

major comments (1)
  1. [Abstract] Abstract and evaluation sections: all quantitative claims (including the ~0.1 dex scatter and twofold improvement) are obtained exclusively by training and testing on Uchuu-UniverseMachine mocks; the central claim of applicability to real clusters therefore rests on the untested assumption that the joint distribution p(phase-space, M_true) in this simulation—including galaxy formation physics, orbital distributions, projection effects, and interloper statistics—matches observations sufficiently closely for the learned corrections to remain valid. No cross-simulation validation or comparison to observed cluster samples is reported.
minor comments (1)
  1. Additional details on training procedure, validation splits, hyperparameter choices, and overfitting diagnostics would strengthen the soundness of the reported performance metrics.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their careful reading and constructive comments on our manuscript. We address the major comment below, acknowledging the scope of our simulation-based study.

read point-by-point responses

  1. Referee: [Abstract] Abstract and evaluation sections: all quantitative claims (including the ~0.1 dex scatter and twofold improvement) are obtained exclusively by training and testing on Uchuu-UniverseMachine mocks; the central claim of applicability to real clusters therefore rests on the untested assumption that the joint distribution p(phase-space, M_true) in this simulation—including galaxy formation physics, orbital distributions, projection effects, and interloper statistics—matches observations sufficiently closely for the learned corrections to remain valid. No cross-simulation validation or comparison to observed cluster samples is reported.

    Authors: We agree that all reported quantitative results, including the scatter reduction to ~0.1 dex and the twofold improvement, are obtained exclusively from training and testing on Uchuu-UniverseMachine mocks. The manuscript does not include cross-simulation validation or direct comparisons to observed cluster samples. The work is presented as a simulation-based inference framework evaluated in controlled mock settings (both interloper-free and cylindrical), with the abstract and methods sections explicitly stating that the model is trained on this simulation. We do not claim that the learned corrections are directly applicable to real clusters; rather, the results demonstrate the information gain achievable within this simulation's joint distribution of phase-space and mass. This is a genuine limitation, as the validity for observations depends on the simulation's fidelity in galaxy formation, orbits, projections, and interlopers—an assumption not tested here. We will revise the abstract, introduction, and conclusions to more explicitly state that the performance metrics are simulation-specific and to emphasize the need for future cross-validation on other simulations and observational comparisons as next steps. revision: partial

Circularity Check

0 steps flagged

No significant circularity; performance gain is empirical evaluation on simulation test set

full rationale

The paper trains a permutation-invariant Deep Sets + NPE model on Uchuu-UniverseMachine mocks to output residual corrections to the classical M-σ relation and then measures the resulting scatter reduction (~0.1 dex, twofold improvement) by direct comparison against M-σ on held-out simulation realizations. This is a standard supervised-learning evaluation against an independent baseline on the same external data distribution; no equation reduces the reported improvement to a quantity defined by the model itself, no self-citation chain is load-bearing for the central claim, and no fitted parameter is renamed as a prediction. The derivation chain remains self-contained within the simulation-based framework.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim depends on the simulation serving as a faithful proxy for real observations; the neural network weights are learned from that simulation but no additional hand-tuned free parameters are stated.

axioms (1)
  • domain assumption The Uchuu-UniverseMachine simulation accurately represents the mapping from observed galaxy kinematics to true halo mass in the real universe.
    Training and performance claims rest on this simulation being representative of nature.

pith-pipeline@v0.9.1-grok · 5747 in / 1208 out tokens · 34394 ms · 2026-06-27T06:14:35.331670+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

80 extracted references · 71 canonical work pages · 30 internal anchors

  1. [1]

    arXiv e-prints , keywords =

    The SRG/eROSITA All-Sky Survey: Cosmology Constraints from Cluster Abundances in the Western Galactic Hemisphere. arXiv e-prints , keywords =. doi:10.48550/arXiv.2402.08458 , archivePrefix =. 2402.08458 , primaryClass =

    work page doi:10.48550/arxiv.2402.08458

  2. [2]

    arXiv e-prints , keywords =

    Predicting dark matter halo masses from simulated galaxy images and environments. arXiv e-prints , keywords =. doi:10.48550/arXiv.2407.13735 , archivePrefix =. 2407.13735 , primaryClass =

    work page doi:10.48550/arxiv.2407.13735

  3. [3]

    arXiv e-prints , keywords =

    Inferring the mass content of galaxy clusters with satellite kinematics and Jeans Anisotropic modeling. arXiv e-prints , keywords =. doi:10.48550/arXiv.2407.11721 , archivePrefix =. 2407.11721 , primaryClass =

    work page doi:10.48550/arxiv.2407.11721

  4. [4]

    arXiv e-prints , keywords =

    Masses of Sunyaev-Zel'dovich Galaxy Clusters Detected by The Atacama Cosmology Telescope: Stacked Lensing Measurements with Subaru HSC Year 3 data. arXiv e-prints , keywords =. doi:10.48550/arXiv.2407.08201 , archivePrefix =. 2407.08201 , primaryClass =

    work page doi:10.48550/arxiv.2407.08201

  5. [5]

    arXiv e-prints , keywords =

    How does the velocity anisotropy of halo stars, dark matter and satellite galaxies depend on host halo properties?. arXiv e-prints , keywords =. doi:10.48550/arXiv.2407.14827 , archivePrefix =. 2407.14827 , primaryClass =

    work page doi:10.48550/arxiv.2407.14827

  6. [6]

    , keywords =

    A Robust and Efficient Deep Learning Method for Dynamical Mass Measurements of Galaxy Clusters. , keywords =. doi:10.3847/1538-4357/ab4f82 , archivePrefix =. 1902.05950 , primaryClass =

    work page doi:10.3847/1538-4357/ab4f82 1902

  7. [7]

    HaloFlow. I. Neural Inference of Halo Mass from Galaxy Photometry and Morphology. , keywords =. doi:10.3847/1538-4357/ad4344 , archivePrefix =. 2310.04503 , primaryClass =

    work page doi:10.3847/1538-4357/ad4344

  8. [8]

    arXiv e-prints , keywords =

    The FLAMINGO Project: An assessment of the systematic errors in the predictions of models for galaxy cluster counts used to infer cosmological parameters. arXiv e-prints , keywords =. doi:10.48550/arXiv.2408.17217 , archivePrefix =. 2408.17217 , primaryClass =

    work page doi:10.48550/arxiv.2408.17217

  9. [9]

    , keywords =

    The Uchuu-universe machine data set: galaxies in and around clusters. , keywords =. doi:10.1093/mnras/stac3514 , archivePrefix =. 2209.12918 , primaryClass =

    work page doi:10.1093/mnras/stac3514

  10. [10]

    The Rockstar Phase-Space Temporal Halo Finder and the Velocity Offsets of Cluster Cores

    The ROCKSTAR Phase-space Temporal Halo Finder and the Velocity Offsets of Cluster Cores. , keywords =. doi:10.1088/0004-637X/762/2/109 , archivePrefix =. 1110.4372 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/762/2/109

  11. [11]

    Astronomy and Computing , keywords =

    C ^ 2 -GAME: Classification of cluster galaxy membership with machine learning. Astronomy and Computing , keywords =. doi:10.1016/j.ascom.2023.100743 , archivePrefix =. 2205.01700 , primaryClass =

    work page doi:10.1016/j.ascom.2023.100743 2023

  12. [12]

    UniverseMachine: The Correlation between Galaxy Growth and Dark Matter Halo Assembly from z=0-10

    UNIVERSEMACHINE: The correlation between galaxy growth and dark matter halo assembly from z = 0-10. , keywords =. doi:10.1093/mnras/stz1182 , archivePrefix =. 1806.07893 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stz1182

  13. [13]

    Variational Inference with Normalizing Flows

    Variational Inference with Normalizing Flows. arXiv e-prints , keywords =. doi:10.48550/arXiv.1505.05770 , archivePrefix =. 1505.05770 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1505.05770

  14. [14]

    Pyro: Deep Universal Probabilistic Programming

    Pyro: Deep Universal Probabilistic Programming. arXiv e-prints , keywords =. doi:10.48550/arXiv.1810.09538 , archivePrefix =. 1810.09538 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1810.09538

  15. [15]

    40th International Conference on Machine Learning , keywords =

    Sampling-Based Accuracy Testing of Posterior Estimators for General Inference. 40th International Conference on Machine Learning , keywords =. doi:10.48550/arXiv.2302.03026 , archivePrefix =. 2302.03026 , primaryClass =

    work page doi:10.48550/arxiv.2302.03026

  16. [16]

    Tracing cosmic evolution with clusters of galaxies , author =. Rev. Mod. Phys. , volume =. 2005 , month =. doi:10.1103/RevModPhys.77.207 , url =

    work page doi:10.1103/revmodphys.77.207 2005

  17. [17]

    Cosmological Parameters from Observations of Galaxy Clusters

    Cosmological Parameters from Observations of Galaxy Clusters. , keywords =. doi:10.1146/annurev-astro-081710-102514 , archivePrefix =. 1103.4829 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1146/annurev-astro-081710-102514

  18. [18]

    Cosmic Visions Dark Energy: Science

    Cosmic Visions Dark Energy: Science. arXiv e-prints , keywords =. doi:10.48550/arXiv.1604.07626 , archivePrefix =. 1604.07626 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1604.07626

  19. [19]

    Constraints on Cosmological Parameters from Future Galaxy Cluster Surveys

    Constraints on Cosmological Parameters from Future Galaxy Cluster Surveys. , keywords =. doi:10.1086/320939 , archivePrefix =. astro-ph/0002336 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/320939

  20. [20]

    C., Edge , A

    The dynamical state and mass-concentration relation of galaxy clusters. , keywords =. doi:10.1111/j.1365-2966.2012.21892.x , archivePrefix =. 1206.1049 , primaryClass =

    work page doi:10.1111/j.1365-2966.2012.21892.x 2012

  21. [21]

    The mass-concentration relation in massive galaxy clusters at redshift ~1

    The mass-concentration relation in massive galaxy clusters at redshift 1. , keywords =. doi:10.1093/mnras/stt1086 , archivePrefix =. 1306.6096 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stt1086

  22. [22]

    The Mass Profile and Accretion History of Cold Dark Matter Halos

    The mass profile and accretion history of cold dark matter haloes. , keywords =. doi:10.1093/mnras/stt526 , archivePrefix =. 1302.0288 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stt526

  23. [23]

    , keywords =

    Observation of the Coma cluster of galaxies with ROSAT during the all-sky-survey. , keywords =

  24. [24]

    and Allen, Steven W

    Mantz, Adam B. and Allen, Steven W. and Morris, R. Glenn and von der Linden, Anja and Applegate, Douglas E. and Kelly, Patrick L. and Burke, David L. and Donovan, David and Ebeling, Harald , title =. Monthly Notices of the Royal Astronomical Society , volume =. 2016 , month =. doi:10.1093/mnras/stw2250 , url =

    work page doi:10.1093/mnras/stw2250 2016

  25. [25]

    Chandra Measurements of a Complete Sample of X-ray Luminous Galaxy Clusters: The Luminosity-Mass Relation

    Chandra measurements of a complete sample of X-ray luminous galaxy clusters: the luminosity-mass relation. , keywords =. doi:10.1093/mnras/stw2621 , archivePrefix =. 1510.04270 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stw2621

  26. [26]

    , keywords =

    Small-Scale Fluctuations of Relic Radiation. , keywords =. doi:10.1007/BF00653471 , adsurl =

    work page doi:10.1007/bf00653471

  27. [27]

    Comments on Astrophysics and Space Physics , keywords =

    The Observations of Relic Radiation as a Test of the Nature of X-Ray Radiation from the Clusters of Galaxies. Comments on Astrophysics and Space Physics , keywords =

  28. [28]

    Chandra Cluster Cosmology Project III: Cosmological Parameter Constraints

    Chandra Cluster Cosmology Project III: Cosmological Parameter Constraints. , keywords =. doi:10.1088/0004-637X/692/2/1060 , archivePrefix =. 0812.2720 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/692/2/1060

  29. [29]

    Constraints on Omega_m, Omega_L, and Sigma_8, from Galaxy Cluster Redshift Distributions

    Constraints on _ m , _ , and _ 8 from Galaxy Cluster Redshift Distributions. , keywords =. doi:10.1086/324309 , archivePrefix =. astro-ph/0105396 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/324309

  30. [30]

    Constraining dark energy with Sunyaev-Zel'dovich cluster surveys

    Constraining Dark Energy with Sunyaev-Zel'dovich Cluster Surveys. , keywords =. doi:10.1103/PhysRevLett.88.231301 , archivePrefix =. astro-ph/0110353 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1103/physrevlett.88.231301

  31. [31]

    Importance of Cluster Structural Evolution in Using X-ray and SZE Galaxy Cluster Surveys to Study Dark Energy

    Importance of Cluster Structural Evolution in Using X-Ray and Sunyaev-Zeldovich Effect Galaxy Cluster Surveys to Study Dark Energy. , keywords =. doi:10.1086/346179 , archivePrefix =. astro-ph/0208002 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/346179

  32. [32]

    , keywords =

    The Atacama Cosmology Telescope: A Catalog of >4000 Sunyaev-Zel dovich Galaxy Clusters. , keywords =. doi:10.3847/1538-4365/abd023 , archivePrefix =. 2009.11043 , primaryClass =

    work page doi:10.3847/1538-4365/abd023 2009

  33. [33]

    Correlations of Richness and Global Properties in Galaxy Clusters

    Correlations of Richness and Global Properties in Galaxy Clusters. , keywords =. doi:10.1086/345929 , archivePrefix =. astro-ph/0211096 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/345929

  34. [34]

    Weighing the Giants - III. Methods and Measurements of Accurate Galaxy Cluster Weak-Lensing Masses

    Weighing the Giants - III. Methods and measurements of accurate galaxy cluster weak-lensing masses. , keywords =. doi:10.1093/mnras/stt2129 , archivePrefix =. 1208.0605 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stt2129

  35. [35]

    Dark Energy Survey Year 1 Results: Weak Lensing Mass Calibration of redMaPPer Galaxy Clusters

    Dark Energy Survey Year 1 results: weak lensing mass calibration of redMaPPer galaxy clusters. , keywords =. doi:10.1093/mnras/sty2711 , archivePrefix =. 1805.00039 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/sty2711

  36. [36]

    Helvetica Physica Acta , year = 1933, month = jan, volume =

    Die Rotverschiebung von extragalaktischen Nebeln. Helvetica Physica Acta , year = 1933, month = jan, volume =

    1933

  37. [37]

    Galaxy Cluster Mass Reconstruction Project: I. Methods and first results on galaxy-based techniques

    Galaxy cluster mass reconstruction project - I. Methods and first results on galaxy-based techniques. , keywords =. doi:10.1093/mnras/stu545 , archivePrefix =. 1403.4610 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stu545

  38. [38]

    and Diaferio, Antonaldo and Rines, Kenneth J

    Geller, Margaret J. and Diaferio, Antonaldo and Rines, Kenneth J. and Serra, Ana Laura , title =. The Astrophysical Journal , abstract =. 2013 , month =. doi:10.1088/0004-637X/764/1/58 , url =

    work page doi:10.1088/0004-637x/764/1/58 2013

  39. [39]

    Dynamical Mass Measurements of Contaminated Galaxy Clusters Using Machine Learning

    Dynamical Mass Measurements of Contaminated Galaxy Clusters Using Machine Learning. , keywords =. doi:10.3847/0004-637X/831/2/135 , archivePrefix =. 1509.05409 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.3847/0004-637x/831/2/135

  40. [40]

    A Deep Learning Approach to Galaxy Cluster X-ray Masses

    A Deep Learning Approach to Galaxy Cluster X-Ray Masses. , keywords =. doi:10.3847/1538-4357/ab14eb , archivePrefix =. 1810.07703 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.3847/1538-4357/ab14eb

  41. [41]

    , keywords =

    Simulation-based inference of dynamical galaxy cluster masses with 3D convolutional neural networks. , keywords =. doi:10.1093/mnras/staa3922 , archivePrefix =. 2009.03340 , primaryClass =

    work page doi:10.1093/mnras/staa3922 2009

  42. [42]

    Mass Estimation of Galaxy Clusters with Deep Learning. I. Sunyaev-Zel'dovich Effect. , keywords =. doi:10.3847/1538-4357/aba694 , archivePrefix =. 2003.06135 , primaryClass =

    work page doi:10.3847/1538-4357/aba694 2003

  43. [43]

    Cosmic Microwave Background Cluster Lensing

    Mass Estimation of Galaxy Clusters with Deep Learning II. Cosmic Microwave Background Cluster Lensing. , keywords =. doi:10.3847/1538-4357/ac32d0 , archivePrefix =. 2005.13985 , primaryClass =

    work page doi:10.3847/1538-4357/ac32d0 2005

  44. [44]

    , keywords =

    Approximate Bayesian Uncertainties on Deep Learning Dynamical Mass Estimates of Galaxy Clusters. , keywords =. doi:10.3847/1538-4357/abd101 , archivePrefix =. 2006.13231 , primaryClass =

    work page doi:10.3847/1538-4357/abd101 2006

  45. [45]

    Nature Astronomy , keywords =

    The dynamical mass of the Coma cluster from deep learning. Nature Astronomy , keywords =. doi:10.1038/s41550-022-01711-1 , archivePrefix =. 2206.14834 , primaryClass =

    work page doi:10.1038/s41550-022-01711-1

  46. [46]

    , keywords =

    Benchmarks and explanations for deep learning estimates of X-ray galaxy cluster masses. , keywords =. doi:10.1093/mnras/stad2005 , archivePrefix =. 2303.00005 , primaryClass =

    work page doi:10.1093/mnras/stad2005

  47. [47]

    , keywords =

    Galaxy cluster mass estimation with deep learning and hydrodynamical simulations. , keywords =. doi:10.1093/mnras/staa3030 , archivePrefix =. 2005.11819 , primaryClass =

    work page doi:10.1093/mnras/staa3030 2005

  48. [48]

    arXiv e-prints , keywords =

    Estimating Dark Matter Halo Masses in Simulated Galaxy Clusters with Graph Neural Networks. arXiv e-prints , keywords =. doi:10.48550/arXiv.2411.12629 , archivePrefix =. 2411.12629 , primaryClass =

    work page doi:10.48550/arxiv.2411.12629

  49. [49]

    arXiv e-prints , keywords =

    HaloFlow II: Robust Galaxy Halo Mass Inference with Domain Adaptation. arXiv e-prints , keywords =. doi:10.48550/arXiv.2603.12380 , archivePrefix =. 2603.12380 , primaryClass =

    work page doi:10.48550/arxiv.2603.12380

  50. [50]

    , keywords =

    Machine-learning Cosmology from Void Properties. , keywords =. doi:10.3847/1538-4357/aceaf6 , archivePrefix =. 2212.06860 , primaryClass =

    work page doi:10.3847/1538-4357/aceaf6

  51. [51]

    Machine Learning: Science and Technology , keywords =

    Predicting the thermal Sunyaev-Zel'dovich field using modular and equivariant set-based neural networks. Machine Learning: Science and Technology , keywords =. doi:10.1088/2632-2153/ac78c2 , archivePrefix =. 2203.00026 , primaryClass =

    work page doi:10.1088/2632-2153/ac78c2

  52. [52]

    , keywords =

    Quijote-PNG: The Information Content of the Halo Mass Function. , keywords =. doi:10.3847/1538-4357/acfe70 , archivePrefix =. 2305.10597 , primaryClass =

    work page doi:10.3847/1538-4357/acfe70

  53. [53]

    , keywords =

    Robust Field-level Likelihood-free Inference with Galaxies. , keywords =. doi:10.3847/1538-4357/acd1e2 , archivePrefix =. 2302.14101 , primaryClass =

    work page doi:10.3847/1538-4357/acd1e2

  54. [54]

    Proceedings of the National Academy of Sciences , volume =

    Kyle Cranmer and Johann Brehmer and Gilles Louppe , title =. Proceedings of the National Academy of Sciences , volume =. 2020 , doi =

    2020

  55. [55]

    Proceedings of the 36th International Conference on Machine Learning , pages =

    Automatic Posterior Transformation for Likelihood-Free Inference , author =. Proceedings of the 36th International Conference on Machine Learning , pages =. 2019 , editor =

    2019

  56. [56]

    Flexible statistical inference for mechanistic models of neural dynamics , url =

    Lueckmann, Jan-Matthis and Goncalves, Pedro J and Bassetto, Giacomo and \". Flexible statistical inference for mechanistic models of neural dynamics , url =. Advances in Neural Information Processing Systems , editor =

  57. [57]

    Fast -free Inference of Simulation Models with Bayesian Conditional Density Estimation , url =

    Papamakarios, George and Murray, Iain , booktitle =. Fast -free Inference of Simulation Models with Bayesian Conditional Density Estimation , url =

  58. [58]

    arXiv e-prints , keywords =

    Neural Spline Flows. arXiv e-prints , keywords =. doi:10.48550/arXiv.1906.04032 , archivePrefix =. 1906.04032 , primaryClass =

    work page doi:10.48550/arxiv.1906.04032 1906

  59. [59]

    arXiv e-prints , keywords =

    Invertible Generative Modeling using Linear Rational Splines. arXiv e-prints , keywords =. doi:10.48550/arXiv.2001.05168 , archivePrefix =. 2001.05168 , primaryClass =

    work page doi:10.48550/arxiv.2001.05168 2001

  60. [60]

    Identification of members in the central and outer regions of galaxy clusters

    Identification of Members in the Central and Outer Regions of Galaxy Clusters. , keywords =. doi:10.1088/0004-637X/768/2/116 , archivePrefix =. 1211.3669 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/768/2/116

  61. [61]

    , keywords =

    A Deep Redshift Survey of the Perseus Cluster (A426): Spatial Distribution and Kinematics of Galaxies. , keywords =. doi:10.3847/1538-4365/ad390d , archivePrefix =. 2403.19307 , primaryClass =

    work page doi:10.3847/1538-4365/ad390d

  62. [62]

    , keywords =

    Dark Energy Survey Year 1 Results: Cosmological constraints from cluster abundances and weak lensing. , keywords =. doi:10.1103/PhysRevD.102.023509 , archivePrefix =. 2002.11124 , primaryClass =

    work page doi:10.1103/physrevd.102.023509 2002

  63. [63]

    Cosmological parameters constraints from galaxy cluster mass function measurements in combination with other cosmological data

    Cosmological parameters constraints from galaxy cluster mass function measurements in combination with other cosmological data. Astronomy Letters , keywords =. doi:10.1134/S1063773712060011 , archivePrefix =. 1202.2889 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1134/s1063773712060011

  64. [64]

    Weighing the Giants IV: Cosmology and Neutrino Mass

    Weighing the giants - IV. Cosmology and neutrino mass. , keywords =. doi:10.1093/mnras/stu2096 , archivePrefix =. 1407.4516 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stu2096

  65. [65]

    New constraints on $f(R)$ gravity from clusters of galaxies

    New constraints on f(R) gravity from clusters of galaxies. , keywords =. doi:10.1103/PhysRevD.92.044009 , archivePrefix =. 1412.0133 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1103/physrevd.92.044009

  66. [66]

    arXiv e-prints , keywords =

    Inferring Halo Mass and Scale Radius of Galaxy Clusters Using Convolutional Neural Networks and Uchuu-UniverseMachine Catalogs. arXiv e-prints , keywords =

  67. [67]

    Zaheer, Manzil and Kottur, Satwik and Ravanbakhsh, Siamak and Poczos, Barnabas and Salakhutdinov, Ruslan and Smola, Alexander , year =. Deep. arXiv e-prints , pages =

  68. [68]

    Machine Learning Techniques for Astrophysics and Cosmology: Simulation-Based Inference

    Machine Learning Techniques for Astrophysics and Cosmology: Simulation-Based Inference. arXiv e-prints , keywords =. doi:10.48550/arXiv.2605.10719 , archivePrefix =. 2605.10719 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2605.10719

  69. [69]

    , keywords =

    The Uchuu simulations: Data Release 1 and dark matter halo concentrations. , keywords =. doi:10.1093/mnras/stab1755 , archivePrefix =. 2007.14720 , primaryClass =

    work page doi:10.1093/mnras/stab1755 2007

  70. [70]

    Constraining neutrino properties with a Euclid-like galaxy cluster survey

    Constraining neutrino properties with a Euclid-like galaxy cluster survey. , keywords =. doi:10.1088/1475-7516/2013/06/020 , archivePrefix =. 1303.4550 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/1475-7516/2013/06/020 2013

  71. [71]

    , keywords =

    The observed growth of massive galaxy clusters - IV. Robust constraints on neutrino properties. , keywords =. doi:10.1111/j.1365-2966.2010.16794.x , archivePrefix =. 0911.1788 , primaryClass =

    work page doi:10.1111/j.1365-2966.2010.16794.x 2010

  72. [72]

    Constraints on Cosmology from the Cosmic Microwave Background Power Spectrum of the 2500 ${\rm deg}^2$ SPT-SZ Survey

    Constraints on Cosmology from the Cosmic Microwave Background Power Spectrum of the 2500 deg ^ 2 SPT-SZ Survey. , keywords =. doi:10.1088/0004-637X/782/2/74 , archivePrefix =. 1212.6267 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/0004-637x/782/2/74

  73. [73]

    Possible indication for non-zero neutrino mass and additional neutrino species from cosmological observations

    Possible indication for non-zero neutrino mass and additional neutrino species from cosmological observations. Astronomy Letters , keywords =. doi:10.1134/S1063773713060030 , archivePrefix =. 1301.4791 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1134/s1063773713060030

  74. [74]

    arXiv e-prints , keywords =

    Improved Cosmological Constraints from SDSS redMaPPer Clusters via X-ray Follow-up of a Complete Subsample of Systems. arXiv e-prints , keywords =. doi:10.48550/arXiv.1910.13548 , archivePrefix =. 1910.13548 , primaryClass =

    work page doi:10.48550/arxiv.1910.13548 1910

  75. [75]

    , keywords =

    Cosmological constraints from the abundance, weak lensing, and clustering of galaxy clusters: Application to the SDSS. , keywords =. doi:10.1051/0004-6361/202348296 , archivePrefix =. 2310.09146 , primaryClass =

    work page doi:10.1051/0004-6361/202348296

  76. [76]

    , keywords =

    Multiprobe cosmology from the abundance of SPT clusters and DES galaxy clustering and weak lensing. , keywords =. doi:10.1103/PhysRevD.111.063533 , archivePrefix =. 2412.07765 , primaryClass =

    work page doi:10.1103/physrevd.111.063533

  77. [77]

    Concentrations of Dark Halos from their Assembly Histories

    Concentrations of Dark Halos from Their Assembly Histories. , keywords =. doi:10.1086/338765 , archivePrefix =. astro-ph/0108151 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1086/338765

  78. [78]

    , keywords =

    The statistics of CDM halo concentrations. , keywords =. doi:10.1111/j.1365-2966.2007.12381.x , archivePrefix =. 0706.2919 , primaryClass =

    work page doi:10.1111/j.1365-2966.2007.12381.x 2007

  79. [79]

    , keywords =

    Dynamical state for 964 galaxy clusters from Chandra X-ray images. , keywords =. doi:10.1093/mnras/staa2363 , archivePrefix =. 2008.01299 , primaryClass =

    work page doi:10.1093/mnras/staa2363 2008

  80. [80]

    On the connection between giant radio halos and cluster mergers

    On the Connection Between Giant Radio Halos and Cluster Mergers. , keywords =. doi:10.1088/2041-8205/721/2/L82 , archivePrefix =. 1008.3624 , primaryClass =

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/2041-8205/721/2/l82 2041