arxiv: 2604.22372 · v2 · submitted 2026-04-24 · 🌌 astro-ph.CO

Recognition: unknown

Testing ΛCDM with ANN-Reconstructed Expansion History from Cosmic Chronometers

Yuki Hashimoto , Kazuharu Bamba , Sanjay Mandal

Authors on Pith no claims yet

Pith reviewed 2026-05-08 09:57 UTC · model grok-4.3

classification 🌌 astro-ph.CO

keywords artificial neural networkcosmic chronometersHubble parameter reconstructionLambda CDM modelnon-parametric methodsexpansion historylate-time cosmologymodel-independent probes

0 comments

The pith

An artificial neural network reconstructs the universe's late-time expansion from chronometer data and matches the standard Lambda CDM model.

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

The paper presents a new artificial neural network framework designed to reconstruct the cosmic expansion history without assuming any particular cosmological model. It trains and validates the network through a three-stage screening pipeline on simulated data before using it on real cosmic chronometer measurements of the Hubble parameter. The resulting H(z) values agree with Lambda CDM predictions inside the current observational errors. This match indicates that the method can reliably extract the expansion history in a model-independent manner from available data.

Core claim

The paper demonstrates that an ANN-based non-parametric reconstruction of H(z) from cosmic chronometer data, after training and validation via a three-stage screening pipeline, produces an expansion history that is consistent with the Lambda CDM model predictions within observational uncertainties, thereby confirming the reliability of the proposed approach for late-time cosmic expansion studies.

What carries the argument

The three-stage ANN screening pipeline that trains and validates a neural network for non-parametric reconstruction of the Hubble parameter H(z) from cosmic chronometer measurements.

If this is right

The framework supplies a model-independent route to test whether the standard cosmological model continues to describe the expansion history as more data arrive.
The same screened ANN procedure can be applied directly to other late-time datasets such as type Ia supernovae or baryon acoustic oscillations.
Consistency with Lambda CDM on current chronometer data implies that any future deviation detected by the method would signal the need for extensions beyond the standard model.
The approach avoids parametric assumptions about dark energy or curvature, allowing cleaner checks on the expansion rate at different redshifts.

Where Pith is reading between the lines

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

The method could be cross-checked against Gaussian-process reconstructions on the same chronometer sample to test whether both non-parametric techniques converge on the same H(z).
If the pipeline remains robust, repeated application to future high-redshift chronometer points might help clarify whether the Hubble tension persists in a model-independent setting.
Extending the screening stages to incorporate additional observables such as the sound horizon scale could turn the framework into a broader test of early- and late-time cosmology.

Load-bearing premise

The three-stage screening pipeline ensures that the neural network reconstruction of H(z) is unbiased and free from artifacts or overfitting when applied to cosmic chronometer data.

What would settle it

A statistically significant departure of the ANN-reconstructed H(z) from the Lambda CDM curve when the same pipeline is run on a new, larger set of cosmic chronometer measurements with smaller uncertainties would falsify the claimed alignment.

Figures

Figures reproduced from arXiv: 2604.22372 by Kazuharu Bamba, Sanjay Mandal, Yuki Hashimoto.

**Figure 1.** Figure 1: FIG. 1. End-to-end pipeline for ANN-based non-parametric r view at source ↗

**Figure 2.** Figure 2: FIG. 2. Representative Stage-2 mock reconstructions for th view at source ↗

**Figure 3.** Figure 3: FIG. 3. Stage-3 reconstruction for the Planck 2018 anchor, view at source ↗

**Figure 4.** Figure 4: FIG. 4. Stage-3 reconstruction for the TRGB anchor, view at source ↗

**Figure 5.** Figure 5: FIG. 5. Stage-3 reconstruction for the SH0ES R21 anchor, imp view at source ↗

read the original abstract

In modern cosmology, the rapid growth of high-precision observational data, along with significant theoretical advances, has intensified the challenge of identifying a robust, model-independent framework to probe the expansion history of the Universe. In this work, we propose a novel artificial neural network (ANN)-based framework for the non-parametric reconstruction of the late-time cosmic expansion. The framework is trained and validated through a three-stage screening pipeline prior to its application to real observational data. As a demonstration of its effectiveness, we reconstruct the Hubble parameter $H(z)$ using the latest cosmic chronometer measurements. Our results show that the reconstructed expansion history aligns with the predictions of the $\Lambda$CDM model within observational uncertainties, thereby supporting the robustness and reliability of the proposed approach.

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.

Desk Editor's Note private letter to a colleague

This paper builds a three-stage ANN screening pipeline to non-parametrically reconstruct H(z) from cosmic chronometers and reports consistency with LambdaCDM within uncertainties.

read the letter

The main thing to know is that the authors train an ANN on cosmic chronometer data through a three-stage screening process and find the reconstructed expansion history sits inside LambdaCDM error bars. The work is a direct, model-independent check on the late-time expansion using real observations rather than simulated ones upfront. What is actually new is the specific three-stage pipeline for training and validation before the real-data step; that adds a practical layer of artifact control that earlier ANN cosmology papers sometimes skipped. The approach does a clean job of keeping the reconstruction step separate from the model comparison, which avoids the usual circularity worry. The data handling looks standard for the field and the claim of alignment is at least internally consistent with the stated method. The soft spots are mostly in the lack of visible quantitative detail in the summary: no reconstruction error bands, no chi-squared or residual statistics, and no clear breakdown of how uncertainties from the small chronometer sample propagate through the network. If those numbers are weak or if the mocks show residual bias, the alignment statement loses force. The dataset size also means any neural net needs unusually careful regularization, and that part needs explicit checks. This is for readers who follow non-parametric methods or machine-learning tools in observational cosmology. It is not a paradigm shift but could serve as a useful cross-check for people testing expansion-history tensions. The paper deserves a serious referee because the core idea is reproducible in principle and the pipeline is a modest but honest extension of existing techniques. I would send it to review with a request for the full validation metrics and error propagation details.

Referee Report

2 major / 2 minor

Summary. The paper proposes a three-stage ANN screening pipeline for non-parametric reconstruction of the late-time Hubble expansion history H(z) from cosmic chronometer data. After training and validation on mocks, the method is applied to real observations, with the central claim that the resulting H(z) reconstruction is consistent with flat ΛCDM predictions within observational uncertainties, thereby validating the framework's robustness for model-independent cosmological tests.

Significance. If the pipeline demonstrably yields an unbiased reconstruction without introducing ΛCDM-aligned artifacts, the work supplies a useful addition to non-parametric tools in cosmology, complementing Gaussian-process or spline methods. The alignment result, once quantified, could help assess whether current cosmic-chronometer data independently support the standard model at z ≲ 2. The screening approach addresses a common concern with neural-network reconstructions and, if reproducible, merits attention in the growing literature on machine-learning cosmology.

major comments (2)

[§4] §4 (Results): The assertion that the reconstructed H(z) 'aligns with the predictions of the ΛCDM model within observational uncertainties' is load-bearing for the central claim yet is presented without quantitative support such as a χ² statistic, mean residual, or point-by-point deviation table relative to a fiducial flat-ΛCDM curve. This omission prevents assessment of whether the agreement is meaningful or merely consistent with large error bars.
[§3] §3 (Methodology, three-stage screening pipeline): The claim that the pipeline produces an unbiased, non-parametric H(z) requires explicit validation that the trained network does not preferentially recover ΛCDM-like behavior when applied to mocks generated from alternative expansion histories (e.g., wCDM or early dark energy). No such cross-model bias test is described, leaving open the possibility that the reported alignment is partly methodological.

minor comments (2)

[Abstract] Abstract: Include at least one concrete quantitative indicator (e.g., number of chronometer points, typical reconstruction uncertainty at z=0.5, or reduced χ²) so readers can immediately gauge the strength of the alignment statement.
[Figures] Figure captions (e.g., Fig. 2 or 3): Ensure error bands on the reconstructed H(z) are clearly distinguished from the ΛCDM reference curve and that the redshift range and data points are labeled.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their constructive and detailed review of our manuscript. We address each major comment below and outline the revisions we will make to strengthen the paper.

read point-by-point responses

Referee: [§4] §4 (Results): The assertion that the reconstructed H(z) 'aligns with the predictions of the ΛCDM model within observational uncertainties' is load-bearing for the central claim yet is presented without quantitative support such as a χ² statistic, mean residual, or point-by-point deviation table relative to a fiducial flat-ΛCDM curve. This omission prevents assessment of whether the agreement is meaningful or merely consistent with large error bars.

Authors: We agree that the current presentation relies on visual consistency within uncertainties without explicit quantitative metrics. While the figures demonstrate that the reconstructed H(z) points lie within the error bars of the fiducial flat-ΛCDM prediction, we did not compute or report a χ² value, mean residuals, or a deviation table. In the revised manuscript we will add these quantitative measures: a χ² statistic for the reconstructed H(z) relative to the best-fit flat-ΛCDM model, the mean residual, and a supplementary table listing point-by-point deviations. This will allow a clearer assessment of the agreement. revision: yes
Referee: [§3] §3 (Methodology, three-stage screening pipeline): The claim that the pipeline produces an unbiased, non-parametric H(z) requires explicit validation that the trained network does not preferentially recover ΛCDM-like behavior when applied to mocks generated from alternative expansion histories (e.g., wCDM or early dark energy). No such cross-model bias test is described, leaving open the possibility that the reported alignment is partly methodological.

Authors: We acknowledge that the existing validation used mocks generated under flat-ΛCDM assumptions to verify recovery of the input history. To directly address potential model-dependent bias, we will generate additional mock cosmic-chronometer datasets from wCDM and early-dark-energy cosmologies, re-apply the three-stage ANN screening pipeline, and quantify any systematic preference for ΛCDM-like reconstructions. The results of these cross-model tests, including any residual biases, will be added to Section 3 of the revised manuscript. revision: yes

Circularity Check

0 steps flagged

Non-parametric reconstruction independent of ΛCDM; alignment is a post-hoc test

full rationale

The paper trains an ANN on mocks via a three-stage pipeline to reconstruct H(z) non-parametrically from cosmic chronometer data alone, without embedding ΛCDM assumptions in the reconstruction step. The comparison to ΛCDM predictions is performed afterward as an external check. No derivation step reduces the final alignment result to the input data or to a self-citation by construction. The central claim therefore remains a genuine test rather than a tautology.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that cosmic chronometer measurements accurately trace the Hubble parameter without significant systematics and that the ANN pipeline generalizes from training to real data without bias. No explicit free parameters or invented entities are named in the abstract.

axioms (1)

domain assumption Cosmic chronometer data provides reliable, unbiased estimates of H(z) at various redshifts.
Invoked implicitly when applying the trained ANN to real observations to claim consistency with ΛCDM.

pith-pipeline@v0.9.0 · 5424 in / 1190 out tokens · 58036 ms · 2026-05-08T09:57:59.924378+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

69 extracted references · 50 canonical work pages · 8 internal anchors

[1]

1 500 0.069 0.053 0.066 0.054 5.561 0.993
[2]

1 500 0.069 0.058 0.067 0.042 5.752 0.995
[3]

1 500 0.065 0.059 0.063 0.044 5.555 0.996
[4]

1 500 0.066 0.063 0.063 0.051 5.402 0.994
[5]

1 500 0.090 0.087 0.087 0.076 5.967 0.989
[6]

The full set o f score statistics is reported in Table IV

1 500 0.130 0.123 0.124 0.109 6.559 0.988 [8, 8] 2 500 0.097 0.087 0.093 0.072 6.946 0.988 [16, 16] 2 500 0.083 0.081 0.078 0.092 6.212 0.992 [32, 32] 2 500 0.103 0.094 0.095 0.160 6.465 0.990 [64, 64] 2 500 0.137 0.141 0.123 0.291 7.116 0.987 [128, 128] 2 500 0.228 0.200 0.198 0.610 9.711 0.977 [256, 256] 2 500 0.269 0.238 0.229 0.800 10.263 0.976 After ...
[7]

For concisio n, the Stage-2 architecture search is summarized only through Table IV; no separate architecture ﬁgure is retained in the revised manuscript

architecture in the Stage-3 results below. For concisio n, the Stage-2 architecture search is summarized only through Table IV; no separate architecture ﬁgure is retained in the revised manuscript. B. Stage 3: reconstruction of the observed CC compilation We next apply the exported Stage-2 conﬁguration to the real C C compilation augmented by one external...
[8]

For each prior choice, the ﬁ nal prediction is constructed from an explicit ensemble of 100 independently initialized training members

network with ELU activation, weighted-L1 loss, zero dro pout, an active sigma head, a learning rate of 10 − 3, a maximum of 800 epochs, no validation split, early-stoppi ng patience 80, and gradient clipping at 5.0. For each prior choice, the ﬁ nal prediction is constructed from an explicit ensemble of 100 independently initialized training members. 16 TA...

2018
[9]

239 for the SH0ES R21 anchor

240 for the TRGB anchor, and 73. 239 for the SH0ES R21 anchor. The prior dependence is therefore explicit and should be regarded as part of the St age-3 setup. At the same time, the observed-space ﬁt diagnostics remain stable across the three runs, with RMSE values between 11. 45 and 14. 18 km s− 1 Mpc− 1 and empirical 68% coverage between 0. 853 and 0. 9...
[10]

Planck 2018 results. VI. Cosmological parameters

N. Aghanim et al. , Astron. Astrophys. 641, A6 (2020), arXiv:1807.06209 [astro-ph.CO]

work page internal anchor Pith review arXiv 2020
[11]

J. A. Frieman, M. S. Turner, and D. Huterer, Ann. Rev. Astr on. Astrophys. 46, 385 (2008), arXiv:0803.0982 [astro-ph]

work page Pith review arXiv 2008
[12]

Padmanabhan, Phys

T. Padmanabhan, Phys. Rept. 380, 235 (2003), arXiv:hep-th/0212290 [hep-th]

work page arXiv 2003
[13]

Data Release 1 of the Dark Energy Spectroscopic Instrument

M. Abdul Karim et al. (DESI), Astron. J. 171, 285 (2026), arXiv:2503.14745 [astro-ph.CO]

work page internal anchor Pith review arXiv 2026
[14]

A. G. Adame et al. , JCAP 2025 (02), 021, arXiv:2404.03002 [astro-ph.CO]

work page internal anchor Pith review arXiv 2025
[15]

Tensions between the Early and the Late Universe

L. Verde, T. Treu, and A. G. Riess, Nat. Astron. 3, 891 (2019), arXiv:1907.10625 [astro-ph.CO]

work page internal anchor Pith review arXiv 2019
[16]

In the Realm of the Hubble tension $-$ a Review of Solutions

E. Di Valentino, O. Mena, S. Pan, L. Visinelli, W. Yang, A. Melchiorri, D. F. Mota, A. G. Riess, and J. Silk, Class. Quant. Grav. 38, 153001 (2021), arXiv:2103.01183 [astro-ph.CO]

work page internal anchor Pith review arXiv 2021
[17]

Perivolaropoulos and F

L. Perivolaropoulos and F. Skara, New Astron. Rev. 95, 101659 (2022), arXiv:2105.05208 [astro-ph.CO]

work page arXiv 2022
[18]

Sch¨ oneberg, G

N. Schöneberg, G. Franco Abellán, A. Pérez Sánchez, S. J. Witte, V. Poulin, and J. Lesgour- gues, Phys. Rept. 984, 1 (2022), arXiv:2107.10291 [astro-ph.CO]

work page arXiv 2022
[19]

Vagnozzi, Universe 9, 393 (2023), arXiv:2308.16628 [astro- ph.CO]

S. Vagnozzi, Universe 9, 393 (2023), arXiv:2308.16628 [astro-ph.CO]

work page arXiv 2023
[20]

A. G. Riess, W. Yuan, L. M. Macri, D. Scolnic, D. Brout, S. Casertano, D. O. Jones, Y. Mu- rakami, G. S. Anand, L. Breuval, T. G. Brink, A. V. Filippenko , S. Hoﬀmann, S. W. Jha, W. D. Kenworthy, J. Mackenty, B. E. Stahl, and W. Zheng, Astrophys . J. Lett. 934, L7 (2022), arXiv:2112.04510 [astro-ph.CO]

work page internal anchor Pith review arXiv 2022
[21]

W. L. Freedman, B. F. Madore, T. J. Hoyt, I. S. Jang, A. J. L ee, and K. A. Owens, Astrophys. J. 985, 203 (2025), [Erratum: Astrophys.J. 993, 252 (2025)], arXiv:2408.06153 [astro-ph.CO]

work page arXiv 2025
[22]

Lei, Z.-F

L. Lei, Z.-F. Wang, T.-L. Wang, Y.-Y. Wang, G.-W. Yuan, W .-L. Lin, and Y.-Z. Fan, Mon. Not. Roy. Astron. Soc. 547, 1 (2026), arXiv:2508.04277 [astro-ph.CO]

work page arXiv 2026
[23]

Constraining Cosmological Parameters Based on Relative Galaxy Ages

R. Jimenez and A. Loeb, Astrophys. J. 573, 37 (2002), arXiv:astro-ph/0106145 [astro-ph]

work page Pith review arXiv 2002
[24]

Constraints on the redshift dependence of the dark energy potential

J. Simon, L. Verde, and R. Jimenez, Phys. Rev. D 71, 123001 (2005), arXiv:astro-ph/0412269 [astro-ph]

work page Pith review arXiv 2005
[25]

Cosmic Chronometers: Constraining the Equation of State of Dark Energy. I: H(z) Measurements

D. Stern, R. Jimenez, L. Verde, M. Kamionkowski, and S. A . Stanford, JCAP 2010 (02), 008, arXiv:0907.3149 [astro-ph.CO]

work page Pith review arXiv 2010
[26]

Improved constraints on the expansion rate of the Universe up to z~1.1 from the spectroscopic evolution of cosmic chronometers

M. Moresco et al. , JCAP 2012 (08), 006, arXiv:1201.3609 [astro-ph.CO]

work page Pith review arXiv 2012
[27]

A 6% measurement of the Hubble parameter at $z\sim0.45$: direct evidence of the epoch of cosmic re-acceleration

M. Moresco et al. , JCAP 2016 (05), 014, arXiv:1601.01701 [astro-ph.CO]

work page Pith review arXiv 2016
[28]

A. L. Ratsimbazafy et al. , Mon. Not. Roy. Astron. Soc. 467, 3239 (2017), arXiv:1702.00418 [astro-ph.CO]

work page Pith review arXiv 2017
[29]

Borghi, M

N. Borghi, M. Moresco, and A. Cimatti, Astrophys. J. Let t. 928, L4 (2022), arXiv:2110.04304 [astro-ph.CO]

work page arXiv 2022
[30]

K. Jiao, N. Borghi, M. Moresco, and T.-J. Zhang, Astroph ys. J. Suppl. 265, 48 (2023), arXiv:2205.05701 [astro-ph.CO]

work page arXiv 2023
[31]

Moresco, R

M. Moresco, R. Jimenez, L. Verde, A. Cimatti, and L. Pozz etti, Astrophys. J. 898, 82 (2020), arXiv:2003.07362 [astro-ph.GA]

work page arXiv 2020
[32]

Moresco et al

M. Moresco et al. , Living Rev. Rel. 25, 6 (2022), arXiv:2201.07241 [astro-ph.CO]

work page arXiv 2022
[33]

Z. f. Wang, L. Lei, and Y. z. Fan, arXiv:2601.07345 [astr o-ph.CO] (2026). 22

work page internal anchor Pith review Pith/arXiv arXiv 2026
[34]

D. M. Scolnic et al. , Astrophys. J. 859, 101 (2018), arXiv:1710.00845 [astro-ph.CO]

work page Pith review arXiv 2018
[35]

The Pantheon+ Analysis: The Full Dataset and Light-Curve Release

D. Scolnic et al. , Astrophys. J. 938, 113 (2022), arXiv:2112.03863 [astro-ph.CO]

work page internal anchor Pith review arXiv 2022
[36]

The 6dF Galaxy Survey: Baryon Acoustic Oscillations and the Local Hubble Constant

F. Beutler et al. , Mon. Not. Roy. Astron. Soc. 416, 3017 (2011), arXiv:1106.3366 [astro-ph.CO]

work page Pith review arXiv 2011
[37]

The clustering of galaxies in the completed SDSS-III Baryon Oscillation Spectroscopic Survey: cosmological analysis of the DR12 galaxy sample

S. Alam et al. , Mon. Not. Roy. Astron. Soc. 470, 2617 (2017), arXiv:1607.03155 [astro-ph.CO]

work page Pith review arXiv 2017
[38]

Alam and others (eBOSS), Phys

S. Alam et al. , Phys. Rev. D 103, 083533 (2021), arXiv:2007.08991 [astro-ph.CO]

work page arXiv 2021
[39]

Shafieloo, Model Independent Reconstruction of the Expansion History of the Universe and the Properties of Dark Energy, Mon

A. Shaﬁeloo, Mon. Not. Roy. Astron. Soc. 380, 1573 (2007), arXiv:astro-ph/0703034 [astro-ph]

work page arXiv 2007
[40]

Reconstruction of dark energy and expansion dynamics using Gaussian processes

M. Seikel, C. Clarkson, and M. Smith, JCAP 2012 (06), 036, arXiv:1204.2832 [astro-ph.CO]

work page Pith review arXiv 2012
[41]

Gaussian processes reconstruction of dark energy from observational data

M.-J. Zhang and H. Li, Eur. Phys. J. C 78, 460 (2018), arXiv:1806.02981 [astro-ph.CO]

work page Pith review arXiv 2018
[42]

Hwang, H

Y. Hwang, H. Noh, and C.-G. Park, JCAP 2023 (04), 071

2023
[43]

Nonparametric Dark Energy Reconstruction from Supernova Data

T. Holsclaw, U. Alam, B. Sansó, H. Lee, K. Heitmann, S. Ha bib, D. Higdon, and D. Bruhwiler, Phys. Rev. Lett. 105, 241302 (2010), arXiv:1011.3079 [astro-ph.CO]

work page Pith review arXiv 2010
[44]

Nonparametric Reconstruction of the Dark Energy Equation of State

T. Holsclaw, U. Alam, B. Sansó, H. Lee, K. Heitmann, S. Ha bib, D. Higdon, and D. Bruhwiler, Phys. Rev. D 82, 103502 (2010), arXiv:1009.5443 [astro-ph.CO]

work page Pith review arXiv 2010
[45]

Nesseris and J

S. Nesseris and J. García-Bellido, JCAP 2012 (11), 033, arXiv:1205.0364 [astro-ph.CO]

work page arXiv 2012
[46]

Wang, X.-J

G.-J. Wang, X.-J. Ma, S.-Y. Li, and J.-Q. Xia, Astrophys . J. Suppl. 246, 13 (2020), arXiv:1910.03636 [astro-ph.CO]

work page arXiv 2020
[47]

Gomez-Vargas, R

I. Gomez-Vargas, R. Medel-Esquivel, R. Garcia-Salced o, and J. A. Vazquez, Eur. Phys. J. C 83, 304 (2023)

2023
[48]

R. Shah, S. Saha, P. Mukherjee, U. Garain, and S. Pal, Ast rophys. J. Suppl. 273, 27 (2024), arXiv:2401.17029 [astro-ph.CO]

work page arXiv 2024
[49]

K. F. Dialektopoulos, J. L. Said, J. Mifsud, J. Sultana, and K. Z. Adami, JCAP 02, 023, arXiv:2111.11462 [astro-ph.CO]

work page arXiv
[50]

Mukherjee, J

P. Mukherjee, J. L. Said, and J. Mifsud, JCAP 12, 029, arXiv:2209.01113 [astro-ph.CO]

work page arXiv
[51]

K. F. Dialektopoulos, P. Mukherjee, J. L. Said, and J. Mi fsud, Eur. Phys. J. C 83, 956 (2023), arXiv:2305.15499 [astro-ph.CO]

work page arXiv 2023
[52]

K. F. Dialektopoulos, P. Mukherjee, J. L. Said, and J. Mi fsud, Phys. Dark Univ. 43, 101383 (2024)

2024
[53]

Abedin, G.-J

M. Abedin, G.-J. Wang, Y.-Z. Ma, and S. Pan, Mon. Not. Roy . Astron. Soc. 540, 2253 (2025)

2025
[54]

Wang, S.-Y

G.-J. Wang, S.-Y. Li, and J.-Q. Xia, Astrophys. J. Suppl . 249, 25 (2020), arXiv:2005.07089 [astro-ph.CO]

work page arXiv 2020
[55]

Wang, Y.-B

Y.-C. Wang, Y.-B. Xie, T.-J. Zhang, H.-C. Huang, T.-T. Z hang, and K. Liu, Astrophys. J. Suppl. 254, 43 (2021), arXiv:2005.10628 [astro-ph.CO]

work page arXiv 2021
[56]

G.-J. Wang, Z. Meng, and J.-Q. Xia, Astrophys. J. Suppl. 268, 7 (2023), arXiv:2306.11102 [astro-ph.CO]

work page arXiv 2023
[57]

Lemos, M

P. Lemos, M. Cranmer, M. Abidi, C. Hahn, M. Eickenberg, E . Massara, D. Yallup, and S. Ho, Mach. Learn. Sci. Technol. 4, 01LT01 (2023)

2023
[58]

Chen, T.-J

J.-F. Chen, T.-J. Zhang, P. He, T. Zhang, and J. Zhang, Eu r. Phys. J. C 85, 1005 (2025)

2025
[59]

Zhang, Y

J.-C. Zhang, Y. Hu, K. Jiao, H.-F. Wang, Y.-B. Xie, B. Yu, L.-L. Zhao, and T.-J. Zhang, Astrophys. J. Suppl. 270, 23 (2024)

2024
[60]

J. Cui, M. Biesiada, A. Liu, C. Wen, T. Liu, and J. Wang, As trophys. J. Suppl. 10.3847/1538-4365/adef3f (2025), arXiv:2504.00392 [ast ro-ph.CO]

work page doi:10.3847/1538-4365/adef3f 2025
[61]

Luo and N

X. Luo and N. Liang, Mon. Not. Roy. Astron. Soc. 542, 1596 (2025)

2025
[62]

Four New Observational $H(z)$ Data From Luminous Red Galaxies of Sloan Digital Sky Survey Data Release Seven

C. Zhang, H. Zhang, S. Yuan, S.-Q. Liu, T.-J. Zhang, and Y .-C. Sun, 23 Res. Astron. Astrophys. 14, 1221 (2014), arXiv:1207.4541 [astro-ph.CO]

work page Pith review arXiv 2014
[63]

Raising the bar: new constraints on the Hubble parameter with cosmic chronometers at z$\sim$2

M. Moresco, Mon. Not. Roy. Astron. Soc. Lett. 450, L16 (2015), arXiv:1503.01116 [astro-ph.CO]

work page Pith review arXiv 2015
[64]

Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)

D.-A. Clevert, T. Unterthiner, and S. Hochreiter, arXi v e-prints , arXiv:1511.07289 (2015), arXiv:1511.07289 [cs.LG]

work page Pith review arXiv 2015
[65]

LeCun, L

Y. LeCun, L. Bottou, G. B. Orr, and K. R. Müller, Eﬃcient b ackprop, in Neural Networks: Tricks of the Trade , edited by G. B. Orr and K.-R. Müller (Springer Berlin Heidelberg, Berlin, Heidelberg, 1998) pp. 9–50

1998
[66]

Elfwing, E

S. Elfwing, E. Uchibe, and K. Doya, Neural Networks 107, 3 (2018), special issue on deep reinforcement learning

2018
[67]

Liu and X

X. Liu and X. Di, IET Computer Vision 15, 136 (2021), arXiv:2003.09855 [cs.LG]

work page arXiv 2021
[68]

Two new diagnostics of dark energy

V. Sahni, A. Shaﬁeloo, and A. A. Starobinsky, Phys. Rev. D 78, 103502 (2008), arXiv:0807.3548 [astro-ph]

work page Pith review arXiv 2008
[69]

W. L. Freedman, B. F. Madore, D. Hatt, T. J. Hoyt, I. S. Jan g, R. L. Beaton, C. R. Burns, M. G. Lee, A. J. Monson, J. R. Neeley, M. M. Phillips, J. A. Rich, and M. Seib- ert, Astrophys. J. 882, 34 (2019), arXiv:1907.05922 [astro-ph.CO]. 24

work page arXiv 2019