Pith. sign in

REVIEW 5 major objections 6 minor 74 references

Machine-assisted Semi-Simulation Model (MSSM): Estimating Galactic Baryonic Properties from their Dark Matter using a Machine Trained on Hydrodynamic Simulations

T0 review · 5 major / 6 minor · reviewed 2026-08-14 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read A machine trained on a hydrodynamic simulation can estimate baryonic galaxy properties from dark-matter halo properties alone, then paint them onto a gigaparsec dark-matter-only volume.

desk verdict Solid, honest engineering paper that improves ML-based halo painting with real technical contributions; the Gpc application is the weak link because the cross-simulation transfer is asserted rather than validated. read the letter →

arxiv 1908.09844 v1 pith:2SNT5GIP submitted 2019-08-26 astro-ph.GA

Yongseok Jo , Ji-hoon Kim This is my paper · ORCID
classification astro-ph.GA
keywords galaxyformationdarkmatterhalosmachinelearninghydrodynamicsimulationssemi-analyticmodelscataloguesextremelyrandomizedtreesIllustrisTNG
topics Dark Matter
open problems Dark Matter
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper claims that the galaxy-halo connection inside a hydrodynamic simulation can be learned as a machine mapping from dark-matter halo properties to baryonic galaxy properties, then applied to a much larger dark-matter-only simulation. If true, this offers a cheap way to build galaxy catalogues at gigaparsec scales, where full hydrodynamic runs are infeasible. The machine-assisted semi-simulation model (MSSM) uses an extremely randomized tree algorithm with three improvements: a logarithmically scaled error function, historical and environmental halo features, and two-stage learning in which predicted stellar magnitudes feed a second model. Trained on IllustrisTNG's $(75\,h^{-1}\,\mathrm{Mpc})^3$ volume, it predicts stellar mass, gas mass, black hole mass, star formation rate, metallicity, and stellar magnitudes with smaller mean binned errors than the baseline; applied to the $(1\,h^{-1}\,\mathrm{Gpc})^3$ MultiDark-Planck run, its galaxy distribution functions are largely compatible with semi-analytic catalogues.

What carries the argument

The load-bearing mechanism is the extremely randomized tree (ERT), an ensemble of decision trees that chooses node splits randomly. Three training choices carry the accuracy gain: (1) a logarithmically scaled error function, so that fractional errors at the low-mass end are not overwhelmed by absolute errors at the high-mass end; (2) additional input features extracted from each halo's merger tree and local environment, such as merger counts, last-major-merger mass ratio, local density, and the semi-potential $\Phi_s=\sum_i M_i/R_i$ within a $(2\,\mathrm{Mpc})^3$ volume; and (3) two-stage learning, in which a first ERT predicts stellar magnitudes from dark-matter features and those predicted magnitudes become inputs to a second ERT that predicts star formation rate. The first machine also outputs feature importances, which show maximum circular velocity dominating at high redshift and halo mass and velocity dispersion taking over by $z=0$.

What would settle it

Run the trained MSSM on IllustrisTNG-Dark, the dark-matter-only counterpart of the training simulation, and compare predicted stellar masses with the actual stellar masses of matched halos in the hydrodynamic run: a systematic bias in the $M_\star$-$M_{\mathrm{halo}}$ relation, or a scatter larger than the intrinsic scatter of that relation, would falsify the transferability assumption. A second check is to compare the two-point galaxy clustering of the MDPL2 catalogue with observed clustering at fixed stellar mass; a significant mismatch would show that the machine's galaxy-halo connection is not faithful enough for large-volume statistics.

Watch

Extended reading notes

Core claim

The paper's central claim is that the baryonic content of a galaxy is predictable, to useful accuracy, from the bulk properties of its dark-matter halo alone, once a machine has been shown enough examples from a hydrodynamic simulation. In the reported test, stellar mass prediction improves from a mean binned error of $0.0018$ (baseline) to $0.0013$, and star formation rate from $1.71$ to $1.00$; the predicted probability distributions for all six baryonic properties move closer to the IllustrisTNG data. When the trained machine is applied to the MultiDark-Planck dark-matter-only simulation, the resulting catalogue's distribution functions are largely compatible with the SAG semi-analytic model, and for black hole mass, star formation rate, and stellar magnitudes it tracks IllustrisTNG more closely than SAG does, which is exactly the behaviour the model was designed for.

Load-bearing premise

The load-bearing premise is that halos in a dark-matter-only simulation match halos in the hydrodynamic simulation closely enough for the learned galaxy-halo mapping to carry over, and that the two simulations' halo mass definitions are compatible after the mass cut; the paper's support for this is a less than 1% agreement in the overall halo mass function, not per-halo comparisons.

Editorial extensions

If this is right

  • Large-scale galaxy catalogues from dark-matter-only runs become a matter of minutes: the machine paints roughly $10^6$ halos in a $(1\,h^{-1}\,\mathrm{Gpc})^3$ volume in tens of minutes, versus weeks for a hydrodynamic simulation.
  • The baryon physics of a specific hydrodynamic simulation can be transplanted onto any sufficiently resolved dark-matter-only simulation without analytic recipes or tuned parameters.
  • Differences between MSSM and SAM catalogues localise where SAM prescriptions deviate from hydrodynamic simulations, pointing to specific subgrid physics to improve.
  • Because the machine reproduces the IllustrisTNG number densities, the resulting catalogue is suited for volume-limited studies such as baryonic acoustic oscillations and large-scale structure statistics.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Pith inference: the two-stage learning trick should generalise; any baryonic property that is accurately predictable from dark matter and strongly correlated with a harder target could be chained as an intermediary, suggesting a searchable design space for future pipelines of this kind.
  • Pith inference: because the transferability assumption is only tested on the halo mass function, a natural next experiment is to run the trained machine on IllustrisTNG-Dark and compare per-halo predictions with the hydrodynamic run; this would isolate transfer error from model error.
  • Pith inference: if the machine's galaxy-halo connection is faithful, the MDPL2 catalogue could be forward-modelled into galaxy clustering, weak-lensing, or CMB-lensing predictions; discrepancies with observations would point to where IllustrisTNG's baryon physics needs revision.
  • Pith inference: the feature-importance analysis suggests that at low redshift halo mass and velocity dispersion carry most of the information; adding merger-orbit or tidal features could broaden the output diversity that the paper identifies as too narrow.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

5 major / 6 minor

Summary. The paper introduces a machine-learning pipeline, MSSM, that uses extremely randomized trees trained on the IllustrisTNG hydrodynamic simulation to predict galaxy baryonic properties (gas mass, stellar mass, black hole mass, star formation rate, metallicity, and stellar magnitudes) from dark-matter halo features. The authors propose three improvements over a baseline model: logarithmic scaling of the error function, addition of historical and environmental halo features, and a two-stage learning scheme that uses predicted stellar magnitudes as an intermediary to predict SFR or stellar mass. They report substantial accuracy gains on a held-out TNG test set, then apply the trained machine to the much larger MultiDark-Planck (MDPL2) DM-only simulation and compare the resulting galaxy catalogue with those from popular semi-analytic models (Sag, Sage, Galacticus), finding 'largely compatible' distributions. The paper also analyzes feature importances and training-set-size requirements, and releases the MSSM galaxy catalogue.

Significance. If the domain-transfer assumption holds, the MSSM provides a fast, flexible way to 'paint' baryonic properties from a high-resolution hydrodynamic simulation onto a very large DM-only volume, which is scientifically valuable for survey-scale predictions. The technical improvements, especially logarithmic error scaling and two-stage learning, are potentially useful to the broader galaxy-formation machine-learning community. The paper is clearly written, gives a detailed description of the pipeline, and makes the catalogue publicly available. However, the headline accuracy claims are measured only inside the TNG training distribution, and the application to MDPL2 rests on a transferability assumption that is validated only via a one-dimensional mass-function comparison. As a result, the main application-phase claim is not yet established with the same rigor as the in-sample accuracy claim.

major comments (5)
  1. [Section 4.1, Section 6.1, Appendix B] The application of the TNG-trained machine to MDPL2 assumes that the per-halo DM feature vectors of the two simulations are drawn from the same distribution. The supporting evidence in Appendix B and Figure B1 is only a comparison of the DM halo mass function, which the authors find to be shifted by less than 1% between IllustrisTNG and IllustrisTNG-Dark. This is a one-dimensional marginal check, while the machine's inputs include halo velocity dispersion, maximum circular velocity, spin, local environmental measures within a (2 Mpc)^3 volume, and three merger-tree quantities (Table 1). Baryonic back-reaction is known to affect halo concentration, Vmax, spin, and subhalo abundance at fixed mass (as the authors themselves cite in Appendix B), and the MDPL2 catalogue is produced with a different halo finder (Rockstar) and a different mass definition than the TNG catalogue (SubFind). A <1% match in the mass function does not constrain these joint feature distributions. The manuscript acknowledges this assumption in Section 6.1 and Appendix B but does not perform the decisive per-halo or joint-distribution test, for example by matching halos between TNG and TNG-Dark and comparing the distribution of the feature vector as a whole. Without such a test, the predicted baryonic properties in the (1 h^-1 Gpc)^3 catalogue may be systematically biased, and the 'largely compatible with SAMs' statement in Section 4.1 is a model-to-model comparison that neither confirms nor refutes such bias.
  2. [Section 3.2.4, Table 2] The best combination of the three proposed improvements is selected separately for each output property by evaluating the MBE and MBSD scores on the same test set that is later used to report the final accuracy values. This is a form of test-set reuse or selection on the test set, which can inflate the reported gains relative to what would be achieved on a truly unseen set. The authors should either use a nested cross-validation procedure (where the test set is used only once, after all model selection is complete) or a separate validation set for selecting among the combinations, and then report the accuracy on the untouched test set. They should also quantify the extent of the inflation, for example by reporting the accuracy of a randomly chosen combination or of the 'all improvements together' model, so that readers can assess the sensitivity to the selection procedure.
  3. [Section 3.1, Section 4.1] The accuracy improvements (e.g., MSE decreasing from 2.0e-2 to 1.9e-4, PCC increasing from 0.971 to 0.987) are measured on a test set drawn from the same TNG100 simulation used for training. This is a legitimate held-out test within the training distribution, but it is interpolation, not out-of-sample prediction. The genuine out-of-sample application, on MDPL2, has no baryonic ground truth, so the comparison with SAMs in Section 4.1 is a comparison between two models and cannot validate the transferred mapping. The abstract's claim of 'significantly increased accuracy compared to prior attempts' should therefore be qualified as accuracy in reproducing the TNG galaxy-halo correlation within the TNG volume, not as accuracy on an independent simulation. The authors should make this distinction explicit in the abstract and conclusions.
  4. [Section 2.5.1, Section 5.2, Figure 8] The training-set sufficiency argument uses learning curves for the baseline model with only three input features (Figure 8 and Section 5.2). The improved model has about fourteen input features (Table 1), including environmental and historical quantities. The authors themselves note in Section 5.2 that if the machine is built with more important input features, a bigger training set may be needed to converge. The current evidence that ~4e4 halos after pruning are sufficient is therefore not directly applicable to the improved model. The authors should provide learning curves for the improved model, or at least for a model with the full feature set, to support the claim that the training set is large enough.
  5. [Section 4.2, Figure 6] The paper reports that the two-dimensional distribution of predicted stellar mass and sSFR is narrower in the MSSM catalogue than in the original IllustrisTNG data (Section 4.2, Figure 6). This narrowing is attributed to underfitting or the limited number of important input features. This is a scientifically relevant limitation because it means the MSSM catalogue does not preserve the full galaxy diversity present in the simulation, and it affects the interpretation of the 'largely compatible' statement in Section 4.1: a distribution that is narrower than the truth can appear compatible in a one-dimensional comparison while being biased in joint and high-order statistics. The authors should quantify this narrowing (e.g., the ratio of standard deviations or a two-sample test) and discuss the practical consequences for downstream scientific applications, such as clustering or abundance matching.
minor comments (6)
  1. [Throughout] There are several typographical errors: 'Becuase' in Section 2.5.1, 'IllutrisTNG' in Section 2.4.1, 'brining' in Section 3.2.1, and 'Feburary' in the header. These should be corrected.
  2. [Section 3.1, Eq. (2)] Equation (2) for MSE appears to be missing the square on the difference term; the text defines it as a mean square error, so the formula should read (1/N) sum (y_pred - y_TNG)^2. If this is a typesetting artifact, please ensure the final version is unambiguous.
  3. [Table 2] The MBSD values for SFR (36.10 and 20.15) are orders of magnitude larger than the MBE values (1.71 and 1.00) in the same table. This seems unusual and should be explained, or the numbers should be checked for a unit or typographical error.
  4. [Section 3.2.3] The two-stage learning scheme uses eight photometric bands as an intermediary. The choice of which bands to use for a given output is described in Appendix A, but in Section 3.2.3 it is stated that for SFR, the g band is used, while for stellar mass, the paper refers to Appendix A without specifying the band set used for the results in Table 2. Please clarify exactly which bands are used for each output property in the 'Best combination' row.
  5. [Figure 4] The residual panels in Figure 4 show the difference between the predicted PDF and the IllustrisTNG PDF, even for the MDPL2 predictions and the SAM catalogue. This is potentially confusing because the residuals do not indicate agreement with MDPL2 ground truth (which does not exist), but rather a comparison to the TNG training data. Please clarify in the caption what the residuals represent.
  6. [Section 6.1] The claim that MSSM 'does not require any recipes with fine-tuned parameters or human bias' is an overstatement, since the choice of input features, the pruning thresholds, and the selection of stellar magnitudes as an intermediary are all human decisions. Please rephrase to acknowledge these choices.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: TNG test-set evaluation is held-out benchmarking and the MDPL2 transfer is an acknowledged assumption, not a by-construction equivalence.

full rationale

The paper's core derivation is a supervised regression from DM features to baryonic properties, trained on a fraction of IllustrisTNG and evaluated on a held-out 20% test set (Sections 2.3.1 and 3.1). This is standard out-of-sample evaluation within one simulation, not a reduction of the reported prediction to the training fit. The accuracy metrics (Eqs. 2-6) compare predictions to unseen TNG test data, so the headline 'significantly increased accuracy ... compared to prior attempts' is an honest benchmark claim about replicating TNG's galaxy-halo correlation. The application to MDPL2 rests on a domain-shift assumption that DM-only halos resemble hydrodynamic-simulation halos. The paper states this assumption explicitly in Section 4 (footnote 15) and Section 6.1, and Appendix B tests only the DM halo mass function (<1% shift in Fig. B1), while acknowledging that baryonic back-reaction may affect internal properties. That is a validation gap or limitation, not circularity: no equation in the paper makes the MDPL2 prediction equal to the training input by construction. The 'largely compatible with SAMs' claim is a model-to-model consistency check, not a claim that SAM agreement is logically entailed by the training procedure. There is no load-bearing self-citation chain and no imported uniqueness theorem; the cited prior work is external and not used to forbid alternatives. Therefore no circular step can be exhibited, and the appropriate finding is no significant circularity.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the transferability of a fitted mapping across simulations and on several preprocessing choices; no new physical entities are introduced.

free parameters (3)
  • ERT hyperparameters (number of trees, max depth, min samples split, etc.) = not reported
    The final model's hyperparameters are not given, so exact reproduction is not possible and the model complexity is a free choice.
  • Best combination of improvements per output property = per-property selection
    Section 3.2.4 selects combinations per output based on test-set performance, effectively a tuning parameter.
  • Pre-processing thresholds (mass cut at 10^9 Msun, environment scale of (2 Mpc)^3, 80% mass ratio) = chosen
    Pre-processing choices in Section 2.5 that affect the training set and features, with no sensitivity analysis.
assumptions (4)
  • domain assumption Baryonic back-reaction does not significantly alter the bulk DM halo properties used as inputs.
    Invoked in Section 4 footnote 15 and Appendix B; supported only by a <1% shift in the DM mass function, not by per-halo comparisons.
  • domain assumption Halo catalogues from SubFind (TNG) and Rockstar (MDPL2) are compatible after pruning to M > 10^9 Msun.
    Implicit in Section 2.5.1; different halo finders and mass definitions are used without correction.
  • ad hoc to paper Stellar magnitudes are a valid intermediary for predicting SFR and stellar mass.
    Section 3.2.3 introduces this as a design choice, motivated by correlation but not by a physical model.
  • domain assumption The test set from TNG100-1 is representative of the target MDPL2 halo population.
    The application phase assumes the trained mapping generalizes across simulations with different resolutions and cosmologies.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Machine-assisted Semi-Simulation Model (MSSM): Estimating Galactic Baryonic Properties from their Dark Matter using a Machine Trained on Hydrodynamic Simulations." pith.science (2026). https://pith.science/paper/2SNT5GIP

@misc{pith2026190809844,
  author       = {Pith},
  title        = {Pith review of: Machine-assisted Semi-Simulation Model (MSSM): Estimating Galactic Baryonic Properties from their Dark Matter using a Machine Trained on Hydrodynamic Simulations},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2SNT5GIP}},
  note         = {Machine review of arXiv:1908.09844}
}
abstract

We present a pipeline to estimate baryonic properties of a galaxy inside a dark matter (DM) halo in DM-only simulations using a machine trained on high-resolution hydrodynamic simulations. As an example, we use the IllustrisTNG hydrodynamic simulation of a $(75 \,\,h^{-1}{\rm Mpc})^3$ volume to train our machine to predict e.g., stellar mass and star formation rate in a galaxy-sized halo based purely on its DM content. An extremely randomized tree (ERT) algorithm is used together with multiple novel improvements we introduce here such as a refined error function in machine training and two-stage learning. Aided by these improvements, our model demonstrates a significantly increased accuracy in predicting baryonic properties compared to prior attempts --- in other words, the machine better mimics IllustrisTNG's galaxy-halo correlation. By applying our machine to the MultiDark-Planck DM-only simulation of a large $(1 \,\,h^{-1}{\rm Gpc})^3$ volume, we then validate the pipeline that rapidly generates a galaxy catalogue from a DM halo catalogue using the correlations the machine found in IllustrisTNG. We also compare our galaxy catalogue with the ones produced by popular semi-analytic models (SAMs). Our so-called machine-assisted semi-simulation model (MSSM) is shown to be largely compatible with SAMs, and may become a promising method to transplant the baryon physics of galaxy-scale hydrodynamic calculations onto a larger-volume DM-only run. We discuss the benefits that machine-based approaches like this entail, as well as suggestions to raise the scientific potential of such approaches.

Figures

Figures reproduced from arXiv: 1908.09844 by the authors.

Figure 1
Figure 1. Flowchart of our machine-assisted semi-simulation model (MSSM). In the learning phase (top panel), we train our machine with a fully hydrodynamic simulation database that contains both dark matter (DM) and baryon data (e.g., IllustrisTNG) to predict the baryonic properties (“output”) based on the DM properties (“input”). In the application phase (bottom panel), by feeding a DM-only N-body simulation (e.g., MultiDark… view at source ↗
Figure 2
Figure 2. Normalized two-dimensional histogram comparing the actual stellar masses of halos in the IllustrisTNG test set, M?,TNG, and the stellar masses predicted from input DM features of the test set, M?,pred. Colors indicate the normalized frequency, nbin = Nbin/Ntot, where Ntot is the total number of halos and Nbin is the number of halos in each two-dimensional bin. Results from two machine learning models are shown: the … view at source ↗
Figure 3
Figure 3. Probability distribution functions Φ (PDFs) of six baryonic properties — gas mass, stellar mass, central black hole mass, star formation rate (SFR), metallicity, and stellar magnitude (g band) — predicted from input DM features in the IllustrisTNG test set. We use two machine learning models to make predictions: the baseline model (blue dot dashed lines; see Section 2.5.2) and our improved model (red solid lines; se… view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Probability distribution functions Φ (PDFs) of six baryonic properties predicted using a DM halo catalogue from the MultiDark-Planck database. Our improved machine trained with IllustrisTNG is applied to a MultiDark-Planck dataset to make predictions (red solid lines; …
Figure 5
Figure 5. Figure 5: Two-dimensional probability distribution of DM halo masses, Mhalo, and predicted stellar masses, M? at z = 0. Colors indicate ρbin = Nbin/(NtotSbin), where Ntot is the total number of halos, Nbin is the number of halos in each two-dimensional bin, and Sbin is the bin a…
Figure 6
Figure 6. Figure 6: Two-dimensional probability distribution of predicted stellar masses, M?, and predicted specific SFRs at z = 0. Colors indicate ρbin = Nbin/(NtotSbin), where Ntot is the total number of halos, Nbin is the number of halos in each two-dimensional bin, and Sbin is the bin…
Figure 7
Figure 7. Figure 7: Relative importances of input features — halo mass, velocity dispersion, maximum circular velocity — when the machine predicts stellar masses (left panel) and central black hole masses (right panel) based only on the three DM features of halos in IllustrisTNG (i.e., ba…
Figure 8
Figure 8. Figure 8: Effect of a training set size on the machine accuracy, Pearson correlation coefficient (PCC), Eq. (3), when the machine predicts various baryonic properties (each of six panels) based on three DM features of halos in IllustrisTNG (i.e., baseline model; see Section 2.5.…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

74 extracted references · 13 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.state := if if FUNCTION not #0 #1 if FUNCTION and 'skip pop #0 if FUNCTION or pop #1...

  2. [2]

    A., 2018, @doi [ ] 10.1093/mnras/sty1169 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.478.3410A 478, 3410

    Agarwal S., Dav \'e R., Bassett B. A., 2018, @doi [ ] 10.1093/mnras/sty1169 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.478.3410A 478, 3410

    doi:10.1093/mnras/sty1169 2018

  3. [4]

    M., 2006, @doi [Reports on Progress in Physics] 10.1088/0034-4885/69/12/R02 , https://ui.adsabs.harvard.edu/\#abs/2006RPPh...69.3101B 69, 3101

    Baugh C. M., 2006, @doi [Reports on Progress in Physics] 10.1088/0034-4885/69/12/R02 , https://ui.adsabs.harvard.edu/\#abs/2006RPPh...69.3101B 69, 3101

    doi:10.1088/0034-4885/69/12/r02 2006

  4. [5]

    S., Wechsler R

    Behroozi P. S., Wechsler R. H., Conroy C., 2013, @doi [ ] 10.1088/0004-637X/770/1/57 , https://ui.adsabs.harvard.edu/\#abs/2013ApJ...770...57B 770, 57

    doi:10.1088/0004-637x/770/1/57 2013

  5. [6]

    arXiv:1806.07893

    Behroozi P., Wechsler R., Hearin A., Conroy C., 2018, arXiv e-prints, https://ui.adsabs.harvard.edu/\#abs/2018arXiv180607893B p. arXiv:1806.07893

    arXiv 2018

  6. [7]

    J., 2010, @doi [ ] 10.1016/j.physrep.2010.06.001 , https://ui.adsabs.harvard.edu/\#abs/2010PhR...495...33B 495, 33

    Benson A. J., 2010, @doi [ ] 10.1016/j.physrep.2010.06.001 , https://ui.adsabs.harvard.edu/\#abs/2010PhR...495...33B 495, 33

    doi:10.1016/j.physrep.2010.06.001 2010

  7. [12]

    S., Dekel A., Kolatt T

    Bullock J. S., Dekel A., Kolatt T. S., Kravtsov A. V., Klypin A. A., Porciani C., Primack J. R., 2001b, @doi [ ] 10.1086/321477 , https://ui.adsabs.harvard.edu/\#abs/2001ApJ...555..240B 555, 240

    doi:10.1086/321477

  8. [13]

    Chua K. T. E., Pillepich A., Vogelsberger M., Hernquist L., 2019, @doi [ ] 10.1093/mnras/sty3531 , https://ui.adsabs.harvard.edu/abs/2019MNRAS.484..476C 484, 476

    doi:10.1093/mnras/sty3531 2019

Show all 74 references
  1. [14]

    Cole S., Lacey C., 1996, @doi [ ] 10.1093/mnras/281.2.716 , https://ui.adsabs.harvard.edu/\#abs/1996MNRAS.281..716C 281, 716

    1996 doi

  2. [15]

    A., et al., 2018, @doi [ ] 10.1093/mnras/sty1131 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.479....2C 479, 2

    Cora S. A., et al., 2018, @doi [ ] 10.1093/mnras/sty1131 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.479....2C 479, 2

    2018 doi

  3. [16]

    J., et al., 2016, @doi [ ] 10.3847/0067-0049/222/2/22 , https://ui.adsabs.harvard.edu/\#abs/2016ApJS..222...22C 222, 22

    Croton D. J., et al., 2016, @doi [ ] 10.3847/0067-0049/222/2/22 , https://ui.adsabs.harvard.edu/\#abs/2016ApJS..222...22C 222, 22

    2016 doi

  4. [17]

    Cui W., Zhang Y., 2017, The Impact of Baryons on the Large-Scale Structure of the Universe . p. 7, @doi 10.5772/68116

    2017 doi

  5. [19]

    F., 2016, @doi [ ] 10.1093/mnras/stw1862 , https://ui.adsabs.harvard.edu/\#abs/2016MNRAS.462.3265D 462, 3265

    Dav \'e R., Thompson R., Hopkins P. F., 2016, @doi [ ] 10.1093/mnras/stw1862 , https://ui.adsabs.harvard.edu/\#abs/2016MNRAS.462.3265D 462, 3265

    2016 doi

  6. [20]

    H., Appleby S., 2019, arXiv e-prints, https://ui.adsabs.harvard.edu/\#abs/2019arXiv190110203D p

    Dav \'e R., Angl \'e s-Alc \'a zar D., Narayanan D., Li Q., Rafieferantsoa M. H., Appleby S., 2019, arXiv e-prints, https://ui.adsabs.harvard.edu/\#abs/2019arXiv190110203D p. arXiv:1901.10203

    2019 arXiv

  7. [21]

    S., White S

    Davis M., Efstathiou G., Frenk C. S., White S. D. M., 1985, @doi [ ] 10.1086/163168 , https://ui.adsabs.harvard.edu/\#abs/1985ApJ...292..371D 292, 371

    1985 doi

  8. [22]

    Dubois Y., et al., 2014, @doi [ ] 10.1093/mnras/stu1227 , https://ui.adsabs.harvard.edu/\#abs/2014MNRAS.444.1453D 444, 1453

    2014 doi

  9. [23]

    R., Schaye J., Kay S

    Duffy A. R., Schaye J., Kay S. T., Dalla Vecchia C., Battye R. A., Booth C. M., 2010, @doi [ ] 10.1111/j.1365-2966.2010.16613.x , https://ui.adsabs.harvard.edu/abs/2010MNRAS.405.2161D 405, 2161

    2010

  10. [24]

    Elbaz D., et al., 2011, @doi [ ] 10.1051/0004-6361/201117239 , https://ui.adsabs.harvard.edu/\#abs/2011A&A...533A.119E 533, A119

    2011 doi

  11. [25]

    Genel S., et al., 2014, @doi [ ] 10.1093/mnras/stu1654 , https://ui.adsabs.harvard.edu/\#abs/2014MNRAS.445..175G 445, 175

    2014 doi

  12. [26]

    Geurts P., Ernst D., Wehenkel L., 2006, @doi [Machine Learning] 10.1007/s10994-006-6226-1 , 63, 3

    2006 doi

  13. [27]

    Heitmann K., et al., 2015, @doi [ ] 10.1088/0067-0049/219/2/34 , https://ui.adsabs.harvard.edu/abs/2015ApJS..219...34H 219, 34

    2015 doi

  14. [28]

    A., Barnes D

    Henson M. A., Barnes D. J., Kay S. T., McCarthy I. G., Schaye J., 2017, @doi [ ] 10.1093/mnras/stw2899 , https://ui.adsabs.harvard.edu/abs/2017MNRAS.465.3361H 465, 3361

    2017 doi

  15. [29]

    Ishiyama T., Enoki M., Kobayashi M. A. R., Makiya R., Nagashima M., Oogi T., 2015, @doi [ ] 10.1093/pasj/psv021 , https://ui.adsabs.harvard.edu/abs/2015PASJ...67...61I 67, 61

    2015 doi

  16. [30]

    Jenkins A., et al., 1998, @doi [ ] 10.1086/305615 , https://ui.adsabs.harvard.edu/\#abs/1998ApJ...499...20J 499, 20

    1998 doi

  17. [31]

    M., Turk M

    Kamdar H. M., Turk M. J., Brunner R. J., 2016a, @doi [ ] 10.1093/mnras/stv2310 , https://ui.adsabs.harvard.edu/\#abs/2016MNRAS.455..642K 455, 642

    doi

  18. [32]

    M., Turk M

    Kamdar H. M., Turk M. J., Brunner R. J., 2016b, @doi [ ] 10.1093/mnras/stv2981 , https://ui.adsabs.harvard.edu/\#abs/2016MNRAS.457.1162K 457, 1162

    doi

  19. [33]

    H., Hernquist L., 1996, @doi [ ] 10.1086/192305 , https://ui.adsabs.harvard.edu/\#abs/1996ApJS..105...19K 105, 19

    Katz N., Weinberg D. H., Hernquist L., 1996, @doi [ ] 10.1086/192305 , https://ui.adsabs.harvard.edu/\#abs/1996ApJS..105...19K 105, 19

    1996 doi

  20. [34]

    V., Khokhlov A

    Klypin A., Gottl \"o ber S., Kravtsov A. V., Khokhlov A. M., 1999, @doi [ ] 10.1086/307122 , https://ui.adsabs.harvard.edu/\#abs/1999ApJ...516..530K 516, 530

    1999 doi

  21. [35]

    A., Trujillo-Gomez S., Primack J., 2011, @doi [ ] 10.1088/0004-637X/740/2/102 , https://ui.adsabs.harvard.edu/abs/2011ApJ...740..102K 740, 102

    Klypin A. A., Trujillo-Gomez S., Primack J., 2011, @doi [ ] 10.1088/0004-637X/740/2/102 , https://ui.adsabs.harvard.edu/abs/2011ApJ...740..102K 740, 102

    2011 doi

  22. [36]

    Klypin A., Yepes G., Gottl \"o ber S., Prada F., He S., 2016, @doi [ ] 10.1093/mnras/stw248 , https://ui.adsabs.harvard.edu/\#abs/2016MNRAS.457.4340K 457, 4340

    2016 doi

  23. [37]

    Knebe A., et al., 2018, @doi [ ] 10.1093/mnras/stx2662 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.474.5206K 474, 5206

    2018 doi

  24. [38]

    Lemson G., Kauffmann G., 1999, @doi [ ] 10.1046/j.1365-8711.1999.02090.x , https://ui.adsabs.harvard.edu/\#abs/1999MNRAS.302..111L 302, 111

    1999

  25. [39]

    Madau P., Diemand J., Kuhlen M., 2008, @doi [ ] 10.1086/587545 , https://ui.adsabs.harvard.edu/\#abs/2008ApJ...679.1260M 679, 1260

    2008 doi

  26. [41]

    Moore B., Ghigna S., Governato F., Lake G., Quinn T., Stadel J., Tozzi P., 1999, @doi [ ] 10.1086/312287 , https://ui.adsabs.harvard.edu/\#abs/1999ApJ...524L..19M 524, L19

    1999 doi

  27. [42]

    P., Naab T., White S

    Moster B. P., Naab T., White S. D. M., 2018, @doi [ ] 10.1093/mnras/sty655 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.477.1822M 477, 1822

    2018 doi

  28. [43]

    P., et al., 2018, @doi [ ] 10.1093/mnras/sty618 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.477.1206N 477, 1206

    Naiman J. P., et al., 2018, @doi [ ] 10.1093/mnras/sty618 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.477.1206N 477, 1206

    2018 doi

  29. [44]

    F., Frenk C

    Navarro J. F., Frenk C. S., White S. D. M., 1997, @doi [ ] 10.1086/304888 , https://ui.adsabs.harvard.edu/\#abs/1997ApJ...490..493N 490, 493

    1997 doi

  30. [45]

    arXiv:1812.05609

    Nelson D., et al., 2018a, arXiv e-prints, https://ui.adsabs.harvard.edu/\#abs/2018arXiv181205609N p. arXiv:1812.05609

    arXiv

  31. [46]

    Nelson D., et al., 2018b, @doi [ ] 10.1093/mnras/stx3040 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.475..624N 475, 624

    doi

  32. [47]

    Pakmor R., Springel V., 2013, @doi [ ] 10.1093/mnras/stt428 , https://ui.adsabs.harvard.edu/\#abs/2013MNRAS.432..176P 432, 176

    2013 doi

  33. [48]

    Pakmor R., Bauer A., Springel V., 2011, @doi [ ] 10.1111/j.1365-2966.2011.19591.x , https://ui.adsabs.harvard.edu/\#abs/2011MNRAS.418.1392P 418, 1392

    2011

  34. [49]

    Pedregosa F., et al., 2011, J. Mach. Learn. Res., 12, 2825

    2011

  35. [50]

    Pillepich A., et al., 2018, @doi [ ] 10.1093/mnras/stx2656 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.473.4077P 473, 4077

    2018 doi

  36. [51]

    Planck Collaboration et al., 2014, @doi [ ] 10.1051/0004-6361/201321591 , https://ui.adsabs.harvard.edu/\#abs/2014A&A...571A..16P 571, A16

    2014 doi

  37. [52]

    Planck Collaboration et al., 2016, @doi [ ] 10.1051/0004-6361/201525830 , https://ui.adsabs.harvard.edu/\#abs/2016A&A...594A..13P 594, A13

    2016 doi

  38. [53]

    A., Simonneau E., Betancort-Rijo J., Patiri S., Gottl \"o ber S., Sanchez-Conde M

    Prada F., Klypin A. A., Simonneau E., Betancort-Rijo J., Patiri S., Gottl \"o ber S., Sanchez-Conde M. A., 2006, @doi [ ] 10.1086/504456 , https://ui.adsabs.harvard.edu/\#abs/2006ApJ...645.1001P 645, 1001

    2006 doi

  39. [54]

    A., Cuesta A

    Prada F., Klypin A. A., Cuesta A. J., Betancort-Rijo J. E., Primack J., 2012, @doi [ ] 10.1111/j.1365-2966.2012.21007.x , https://ui.adsabs.harvard.edu/abs/2012MNRAS.423.3018P 423, 3018

    2012

  40. [55]

    Riebe K., et al., 2013, @doi [Astronomische Nachrichten] 10.1002/asna.201211900 , https://ui.adsabs.harvard.edu/\#abs/2013AN....334..691R 334, 691

    2013 doi

  41. [56]

    Rodriguez-Gomez V., et al., 2015, @doi [ ] 10.1093/mnras/stv264 , https://ui.adsabs.harvard.edu/\#abs/2015MNRAS.449...49R 449, 49

    2015 doi

  42. [57]

    Rodr \' guez-Puebla A., Behroozi P., Primack J., Klypin A., Lee C., Hellinger D., 2016, @doi [ ] 10.1093/mnras/stw1705 , https://ui.adsabs.harvard.edu/abs/2016MNRAS.462..893R 462, 893

    2016 doi

  43. [58]

    R., Avila-Reese V., Faber S

    Rodr \' guez-Puebla A., Primack J. R., Avila-Reese V., Faber S. M., 2017, @doi [ ] 10.1093/mnras/stx1172 , https://ui.adsabs.harvard.edu/\#abs/2017MNRAS.470..651R 470, 651

    2017 doi

  44. [59]

    Salim S., et al., 2016, @doi [ ] 10.3847/0067-0049/227/1/2 , https://ui.adsabs.harvard.edu/abs/2016ApJS..227....2S 227, 2

    2016 doi

  45. [60]

    S., Crain R

    Sawala T., Frenk C. S., Crain R. A., Jenkins A., Schaye J., Theuns T., Zavala J., 2013, @doi [ ] 10.1093/mnras/stt259 , https://ui.adsabs.harvard.edu/abs/2013MNRAS.431.1366S 431, 1366

    2013 doi

  46. [61]

    Schaye J., Dalla Vecchia C., 2008, @doi [ ] 10.1111/j.1365-2966.2007.12639.x , https://ui.adsabs.harvard.edu/\#abs/2008MNRAS.383.1210S 383, 1210

    2008

  47. [62]

    Schaye J., et al., 2015, @doi [ ] 10.1093/mnras/stu2058 , https://ui.adsabs.harvard.edu/\#abs/2015MNRAS.446..521S 446, 521

    2015 doi

  48. [63]

    A., 2012, @doi [Research in Astronomy and Astrophysics] 10.1088/1674-4527/12/8/004 , https://ui.adsabs.harvard.edu/\#abs/2012RAA....12..917S 12, 917

    Silk J., Mamon G. A., 2012, @doi [Research in Astronomy and Astrophysics] 10.1088/1674-4527/12/8/004 , https://ui.adsabs.harvard.edu/\#abs/2012RAA....12..917S 12, 917

    2012 doi

  49. [64]

    W., Warren M

    Skillman S. W., Warren M. S., Turk M. J., Wechsler R. H., Holz D. E., Sutter P. M., 2014, arXiv e-prints, https://ui.adsabs.harvard.edu/abs/2014arXiv1407.2600S p. arXiv:1407.2600

    2014 arXiv

  50. [68]

    Springel V., Hernquist L., 2003b, @doi [ ] 10.1046/j.1365-8711.2003.06207.x , https://ui.adsabs.harvard.edu/\#abs/2003MNRAS.339..312S 339, 312

    2003

  51. [69]

    Springel V., White S. D. M., Tormen G., Kauffmann G., 2001, @doi [ ] 10.1046/j.1365-8711.2001.04912.x , https://ui.adsabs.harvard.edu/\#abs/2001MNRAS.328..726S 328, 726

    2001

  52. [70]

    Springel V., Di Matteo T., Hernquist L., 2005a, @doi [ ] 10.1111/j.1365-2966.2005.09238.x , https://ui.adsabs.harvard.edu/\#abs/2005MNRAS.361..776S 361, 776

    2005

  53. [71]

    Springel V., et al., 2005b, @doi [ ] 10.1038/nature03597 , https://ui.adsabs.harvard.edu/\#abs/2005Natur.435..629S 435, 629

    doi

  54. [73]

    Springer International Publishing, pp 21--36

    Springel V., et al., 2018, in High Performance Computing in Science and Engineering ' 17. Springer International Publishing, pp 21--36

    2018

  55. [74]

    R., Pontzen A., Anderson L., Bellovary J., 2017, @doi [ ] 10.1093/mnras/stx1160 , https://ui.adsabs.harvard.edu/\#abs/2017MNRAS.470.1121T 470, 1121

    Tremmel M., Karcher M., Governato F., Volonteri M., Quinn T. R., Pontzen A., Anderson L., Bellovary J., 2017, @doi [ ] 10.1093/mnras/stx1160 , https://ui.adsabs.harvard.edu/\#abs/2017MNRAS.470.1121T 470, 1121

    2017 doi

  56. [75]

    Vogelsberger M., et al., 2014a, @doi [ ] 10.1093/mnras/stu1536 , https://ui.adsabs.harvard.edu/\#abs/2014MNRAS.444.1518V 444, 1518

    doi

  57. [76]

    Vogelsberger M., et al., 2014b, @doi [ ] 10.1038/nature13316 , https://ui.adsabs.harvard.edu/\#abs/2014Natur.509..177V 509, 177

    doi

  58. [77]

    A., Iliev I

    Watson W. A., Iliev I. T., D'Aloisio A., Knebe A., Shapiro P. R., Yepes G., 2013, @doi [ ] 10.1093/mnras/stt791 , https://ui.adsabs.harvard.edu/abs/2013MNRAS.433.1230W 433, 1230

    2013 doi

  59. [78]

    Weinberger R., et al., 2017, @doi [ ] 10.1093/mnras/stw2944 , https://ui.adsabs.harvard.edu/\#abs/2017MNRAS.465.3291W 465, 3291

    2017 doi

  60. [79]

    Weinberger R., et al., 2018, @doi [ ] 10.1093/mnras/sty1733 , https://ui.adsabs.harvard.edu/\#abs/2018MNRAS.479.4056W 479, 4056

    2018 doi

  61. [80]

    White S. D. M., Frenk C. S., Davis M., Efstathiou G., 1987a, @doi [ ] 10.1086/164990 , https://ui.adsabs.harvard.edu/\#abs/1987ApJ...313..505W 313, 505

    doi

  62. [81]

    White S. D. M., Davis M., Efstathiou G., Frenk C. S., 1987b, @doi [ ] 10.1038/330451a0 , https://ui.adsabs.harvard.edu/\#abs/1987Natur.330..451W 330, 451

    doi

  63. [82]

    Wiersma R. P. C., Schaye J., Smith B. D., 2009a, @doi [ ] 10.1111/j.1365-2966.2008.14191.x , https://ui.adsabs.harvard.edu/\#abs/2009MNRAS.393...99W 393, 99

    2008

  64. [83]

    Wiersma R. P. C., Schaye J., Theuns T., Dalla Vecchia C., Tornatore L., 2009b, @doi [ ] 10.1111/j.1365-2966.2009.15331.x , https://ui.adsabs.harvard.edu/\#abs/2009MNRAS.399..574W 399, 574

    2009

  65. [84]

    arXiv:1902.05965

    Zhang X., Wang Y., Zhang W., Sun Y., He S., Contardo G., Villaescusa-Navarro F., Ho S., 2019, arXiv e-prints, https://ui.adsabs.harvard.edu/\#abs/2019arXiv190205965Z p. arXiv:1902.05965

    2019 arXiv

  66. [85]

    write newline

    " write newline "" before.all 'output.state := FUNCTION fin.entry write newline FUNCTION new.block output.state before.all = 'skip after.block 'output.state := if FUNCTION new.sentence output.state after.block = 'skip output.state before.all = 'skip after.sentence 'output.stat...

Pith tools

Reviewed August 14, 2026 · model on record in the stance chip above.