pith. sign in

arxiv: 2606.24383 · v1 · pith:LPKX2YIGnew · submitted 2026-06-23 · 🌌 astro-ph.SR · astro-ph.GA· astro-ph.IM

Evaluating the Sensitivity of the Age Inferences of Red Giant Stars to Machine Learning Methodology

Jamie Tayar , Carli Mankowski , Lara Tunca , Dante Jordan , Mia Severino , Sydney McArthur , Zeina Benton , Sophia Armstrong
show 17 more authors
Alexa Leddy Emily Bower Zabdiel Sanchez Colin Avery Emily Cummings Joshua Donley Rachel Freeman David R. Fulcher Vanessa Hervie James Ivey Hyde Kenney William MacMillan Jake Mahoney Eve Maramba Erin Philip Yazmeen Simpson Ethan Strojie
This is my paper

Pith reviewed 2026-06-25 22:42 UTC · model grok-4.3

classification 🌌 astro-ph.SR astro-ph.GAastro-ph.IM
keywords stellar agesred giant starsmachine learningneural networkstraining setMilky Wayage inferencespectroscopy
0
0 comments X

The pith

Ages for red giant stars inferred by machine learning stay mostly stable when model architecture or hyperparameters change, but shift when the training set changes.

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

The paper tests how much the machine-learning ages of 351995 Milky Way stars depend on choices of neural-network settings, model type, and training data. It finds that ages remain largely unchanged across different hyperparameters and architectures, yet vary more when the training set is swapped. The effect is strongest for the oldest, coolest, and lowest-metallicity stars. This matters because stellar ages are used to reconstruct the galaxy's assembly history, so identifying the dominant source of uncertainty helps decide where to invest effort next. The work concludes that even basic neural networks work adequately once a good training set exists.

Core claim

The resulting ages are generally insensitive to the neural network hyperparameters or the machine learning architecture, but are somewhat sensitive to the training set chosen. Ages for the oldest, coolest, and lowest metallicity stars in the sample are most sensitive to the methodology used and the training set chosen.

What carries the argument

Sensitivity tests that vary neural-network hyperparameters, model architecture, and training-set selection on Milky Way Mapper Data Release 19 stars with known ages.

If this is right

  • Even simple neural-network models are sufficient for accurate age inference from spectroscopic data.
  • Expanding the size and diversity of available training sets will be the main route to reliable ages for the full galactic population.
  • Catalogs of ages for the oldest, coolest, and lowest-metallicity stars should carry extra uncertainty flags tied to training-set choice.

Where Pith is reading between the lines

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

  • If label quality varies across training sets, then part of the reported sensitivity may actually trace back to label errors rather than model choices.
  • Age maps of the Milky Way could be made more robust by reporting an ensemble of ages drawn from multiple training sets instead of a single value.
  • Future work might test whether weighting training stars by temperature or metallicity reduces the sensitivity observed in the oldest objects.

Load-bearing premise

Changes seen when swapping training sets are assumed to measure methodological sensitivity rather than differences in the quality or biases of the known-age labels themselves.

What would settle it

Retraining the same models on an independent catalog of stars whose ages were measured by a completely different method (for example, asteroseismology) and checking whether the oldest stars still show large age shifts would test the claim.

Figures

Figures reproduced from arXiv: 2606.24383 by Alexa Leddy, Carli Mankowski, Colin Avery, Dante Jordan, David R. Fulcher, Emily Bower, Emily Cummings, Erin Philip, Ethan Strojie, Eve Maramba, Hyde Kenney, Jake Mahoney, James Ivey, Jamie Tayar, Joshua Donley, Lara Tunca, Mia Severino, Rachel Freeman, Sophia Armstrong, Sydney McArthur, Vanessa Hervie, William MacMillan, Yazmeen Simpson, Zabdiel Sanchez, Zeina Benton.

Figure 1
Figure 1. Figure 1: General properties of the MWM DR19 sample. Top left: [𝛼/Fe]-[Fe/H] relation, color-coded by the number of stars per bin. Top right: [Fe/H] map in cylindrical coordinates (ZGal versus RGal). Bottom left: [𝛼/M]-[C/N] relation, color-coded by log g. Bottom right: Kiel diagram (log(g) versus Teff), color-coded by [Fe/H]. scales with the surface gravity, and the large frequency sep￾aration, Δ𝜈, which scales wit… view at source ↗
Figure 2
Figure 2. Figure 2: Kiel diagrams of our training, calibration, and comparison samples, color coded by metallicity. The grey box indicates the regime of interest, defined by the training data, for our analysis compared to each sample. values. This provides us a list of 637 giants in 94 open clus￾ters that we can use to evaluate our age results. In order to broaden the age and metallicity ranges of our calibrators, we also inc… view at source ↗
Figure 3
Figure 3. Figure 3: Top: Recovery plot for our preferred base case neural network, Model 2 trained on APOKASC-3 data. Points indicate the predicted versus seismic ages individual stars in the test set, with the black line representing agreement and the dotted red lines marking offsets of 30 percent. Bottom: Loss plot for Model 2 trained on APOKASC-3 data for both the training (blue) and testing (orange) sets. While loss conti… view at source ↗
Figure 5
Figure 5. Figure 5: Demonstration of the impact of changing the training dataset to APOKASC-2 (top, red), APO-K2 (middle, teal), or TESS (bottom, purple) compared to the base case ages inferred by training on APOKASC-3 (blue). When computing the mean and standard deviation for each sample, we have removed significant outliers (|ΔAge| > 5 Gyr). 3.2.1. APO-K2 We have done some additional exploration with the APO￾K2 data to try … view at source ↗
Figure 4
Figure 4. Figure 4: (top) and a more generalized comparison of a set with varied hyperparameters to our preferred, well-trained model ( [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 7
Figure 7. Figure 7: Left: Comparison of our predicted ages to literature ages inferred using more sophisticated machine learning techniques (top, variational encoder-decoder, H. W. Leung et al. 2023; second from top, Bayesian Convolutional Neural Networks, J. T. Mackereth et al. 2019; bottom two rows, normalizing flows, A. Stone-Martinez et al. 2025) but similar training sets (see text). Right: Histograms, colored by the trai… view at source ↗
Figure 6
Figure 6. Figure 6: Residuals between the predictions of the network trained on APOKASC-3 and the network trained on APO-K2 in a Kiel diagram (top), chemical space (middle), and age-metallicity space (bottom). We mark with a gray box the location in each plot of the majority of the APO-K2 training data, and find significant offsets in the results (identified as bright red or bright blue points) far from this region. 3.3. Comp… view at source ↗
Figure 8
Figure 8. Figure 8: Differences between the ages predicted from networks trained on each sample (from top: APOKASC-2, red; APOKASC-3, blue; APO-K2, teal; TESS, purple) and age of the clusters for each member of an open (squares and opaque histograms) or globular (circles and translucent histograms) cluster. We also show the com￾parison between the published A. Stone-Martinez et al. (2025) ages and the cluster ages in the bott… view at source ↗
Figure 10
Figure 10. Figure 10: Offsets between the age predictions of models trained on APOKASC-2 (top, red), APO-K2 (middle, teal), and TESS (bottom, purple) compared to our baseline model trained on APOKASC-3 as a function of temperature (left), surface gravity (middle), and metallicity (right). For each plot, we have marked a running median (solid line) and uncertainty band (16-84th percentile) to emphasize regimes where predictions… view at source ↗
Figure 9
Figure 9. Figure 9: Offsets between the ages predicted by our networks and the seismic ages for values that were not included in the network’s train￾ing set (colors: APOKASC-2, red; APOKASC-3, blue; APO-K2, teal; TESS, purple). Offsets between the measured and predicted values are generally small for stars younger than 9 Gyr. some efforts to directly infer asteroseismic ages from globular cluster stars (e.g. M. Tailo et al. 2… view at source ↗
Figure 11
Figure 11. Figure 11: Average fractional variation in age between models trained on the APOKASC-2, APOKASC-3, and TESS samples as a function of galactocentric radius (left), metallicity (center), and 𝛼-element enhancement (right). In each panel, the color indicates the density of points, with darker regions having more points, and we have shown the running median as a red line. comparison, but we find the results inferred from… view at source ↗
Figure 12
Figure 12. Figure 12: Absolute (top) and fraction (bottom) scatter in ages predicted by models using different training sets (APOKASC-2, APOKASC-3, or TESS) for stars in our sample as a function of Kiel diagram position (left), composition (middle), or age and metallicity (right). Dark colors indicate ages that are consistently inferred while bright colors indicate significant systematic uncertainties in age. High- Low- metal-… view at source ↗
Figure 13
Figure 13. Figure 13: Violin plots showing the distribution of fractional un￾certainties in age predicted by networks using different training sets (APOKASC-2, APOKASC-3, TESS) for subsets of our data. For each subset the average is this a mean or a median value is marked with a black line. using different methods will depend sensitively on the part of parameter space. We further break down these offsets as a function of popul… view at source ↗
Figure 14
Figure 14. Figure 14: Kiel diagrams with each star color-coded by age for stars in each metallicity bin from the most metal poor (top left) to the most metal rich (lower right). In each case, old stars are marked as bright points and young stars are marked as dark points. In each panel, the expected trends are seen in age for both the shell-hydrogen-burning first-ascent red giant branch stars and the core-helium-burning clump … view at source ↗
Figure 15
Figure 15. Figure 15: [𝛼/M] versus [M/H] plots for each age bin in our sample from the youngest (upper left) to the oldest (lower row, second from the right) showing the evolution of chemistry over time in the galaxy, with the number of stars at each composition indicated by color. In the lower right panel, we show the evolution of the 𝛼-element enhancement with age in the sample, with a running median (solid line) and uncerta… view at source ↗
Figure 16
Figure 16. Figure 16: Similar to [PITH_FULL_IMAGE:figures/full_fig_p013_16.png] view at source ↗
Figure 17
Figure 17. Figure 17: Galactic height versus galactocentric radius plots for our stars, broken up into bins of age from the youngest (upper left) to the oldest (lower panel, second from right). In each panel, stars are color coded by their metallicity, indicating the evolution of chemistry with position and time. In the lower right panel, we show the evolution of the average height above the plane in our sample with age, with … view at source ↗
read the original abstract

Stellar ages are vital for understanding the formation of our galaxy, but they are among the most challenging parameters to measure. Many authors address this by using machine learning models trained on stars of known age. Here we used data for 351,995 stars from Milky Way Mapper Data Release 19 to explore the sensitivity of the inferred ages to 1) neural network hyperparameters, 2) machine learning architecture, and 3) training set. We find that the resulting ages are generally insensitive to the neural network hyperparameters or the machine learning architecture, but are somewhat sensitive to the training set chosen. We also find that ages for the oldest, coolest, and lowest metallicity stars in the sample are most sensitive to the methodology used and the training set chosen. In general, our analysis suggests that even simple neural network models are sufficient for accurate age inference, but future work expanding the available training sets will be an important component of predicting reliable ages for the full galactic population.

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 / 0 minor

Summary. The manuscript reports an empirical sensitivity analysis of machine-learning-based age inferences for 351,995 red giant stars from Milky Way Mapper DR19. The central claims are that the resulting ages are generally insensitive to neural-network hyperparameters and architecture choices, but are somewhat sensitive to the training set selected, with the oldest, coolest, and lowest-metallicity stars showing the largest variations; the authors conclude that even simple networks suffice and that expanding training sets is the priority for reliable galactic ages.

Significance. If the results are robust, the work provides a practical benchmark showing that model architecture is not the dominant uncertainty source in ML age inference, directing community effort toward training-set expansion and label quality. This has direct implications for large-scale galactic archaeology catalogs that rely on such ages.

major comments (1)
  1. [Abstract; training-set experiments section] Abstract and the section describing the training-set experiments: the headline claim that ages are 'somewhat sensitive to the training set chosen' (while insensitive to hyperparameters/architecture) is load-bearing for the interpretation that this sensitivity is a methodological property. The analysis assumes that differences across training sets primarily reflect statistical coverage or selection rather than systematic differences in the accuracy, bias, or selection function of the 'known ages' used as regression targets. Without explicit cross-set label-consistency checks (e.g., direct comparison of label distributions, metallicity-dependent offsets, or propagated uncertainties), the observed sensitivity cannot be unambiguously attributed to the ML pipeline rather than inconsistent supervision.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their constructive and detailed review. The major comment identifies a potential ambiguity in attributing training-set sensitivity to the ML pipeline. We address this point directly below and will incorporate clarifications and additional checks in the revised manuscript.

read point-by-point responses

  1. Referee: [Abstract; training-set experiments section] Abstract and the section describing the training-set experiments: the headline claim that ages are 'somewhat sensitive to the training set chosen' (while insensitive to hyperparameters/architecture) is load-bearing for the interpretation that this sensitivity is a methodological property. The analysis assumes that differences across training sets primarily reflect statistical coverage or selection rather than systematic differences in the accuracy, bias, or selection function of the 'known ages' used as regression targets. Without explicit cross-set label-consistency checks (e.g., direct comparison of label distributions, metallicity-dependent offsets, or propagated uncertainties), the observed sensitivity cannot be unambiguously attributed to the ML pipeline rather than inconsistent supervision.

    Authors: We appreciate the referee's emphasis on unambiguous attribution. All training sets were constructed as subsets or selection variants drawn from the identical parent catalog of 351,995 MWM DR19 red giants whose ages were determined by the same underlying method; the labels are therefore drawn from a single, internally consistent source. Observed differences thus arise from variations in statistical coverage, parameter-space sampling, and selection functions rather than from label systematics between independent sources. That said, we agree that explicit verification strengthens the claim. In the revision we will add (i) direct comparisons of the age, [Fe/H], and Teff distributions across the training sets and (ii) checks for metallicity-dependent offsets between the label values themselves. These additions will be placed in the training-set experiments section and referenced in the abstract. We note that full propagation of label uncertainties would require additional external data not available in the current catalog; we will instead discuss this as a limitation and a target for future work. revision: yes

Circularity Check

0 steps flagged

Empirical sensitivity study contains no derivation chain or self-referential steps

full rationale

The paper performs an empirical comparison of neural network age inferences across hyperparameter choices, architectures, and training sets on a fixed dataset of 351995 stars. No equations, first-principles derivations, fitted parameters renamed as predictions, or uniqueness theorems appear in the provided text. All reported sensitivities are direct observational outcomes of retraining and evaluating models; none reduce by construction to the inputs via self-definition or self-citation. The central claim (ages somewhat sensitive to training set) is a measured difference between runs and does not invoke any load-bearing self-citation or ansatz. This is a standard empirical ablation study whose results stand or fall on the data and code rather than on any circular logical step.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Abstract-only review yields minimal ledger entries; the central claim rests on the domain assumption that training labels are sufficiently accurate to isolate methodological effects.

axioms (1)
  • domain assumption Machine learning models trained on stars with known ages can produce usable age inferences for other stars
    Invoked throughout the abstract as the basis for the entire sensitivity experiment.

pith-pipeline@v0.9.1-grok · 5792 in / 1240 out tokens · 26333 ms · 2026-06-25T22:42:51.709533+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

136 extracted references · 129 canonical work pages · 5 internal anchors

  1. [1]

    , keywords =

    Abadi, M., Agarwal, A., Barham, P., et al. 2015, https://www.tensorflow.org/ Abdurro’uf, Accetta, K., Aerts, C., et al. 2022, ApJS, 259, 35, doi: 10.3847/1538-4365/ac4414

    work page doi:10.3847/1538-4365/ac4414 2015

  2. [2]

    ApJS , volume =

    Abolfathi, B., Aguado, D. S., Aguilar, G., et al. 2018, ApJS, 235, 42, doi: 10.3847/1538-4365/aa9e8a

    work page doi:10.3847/1538-4365/aa9e8a 2018

  3. [3]

    Agarap, A. F. 2018, arXiv e-prints, arXiv:1803.08375, doi: 10.48550/arXiv.1803.08375

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1803.08375 2018

  4. [4]

    The Eighteenth Data Release of the Sloan Digital Sky Surveys: Targeting and First Spectra from SDSS-V , volume =

    Almeida, A., Anderson, S. F., Argudo-Fern´andez, M., et al. 2023, ApJS, 267, 44, doi: 10.3847/1538-4365/acda98

    work page doi:10.3847/1538-4365/acda98 2023

  5. [5]

    2023, A&A, 678, A158, doi: 10.1051/0004-6361/202346666

    Anders, F., Gispert, P., Ratcliffe, B., et al. 2023, A&A, 678, A158, doi: 10.1051/0004-6361/202346666

    work page doi:10.1051/0004-6361/202346666 2023

  6. [6]

    2023, The Astrophysical Journal Supplement Series, 267, 8, doi: 10.3847/1538-4365/acd53e

    Andrae, R., Rix, H.-W., & Chandra, V. 2023, The Astrophysical Journal Supplement Series, 267, 8, doi: 10.3847/1538-4365/acd53e

    work page doi:10.3847/1538-4365/acd53e 2023

  7. [7]

    H., Weinberg, D

    Andrews, B. H., Weinberg, D. H., Sch¨onrich, R., & Johnson, J. A. 2017, ApJ, 835, 224, doi: 10.3847/1538-4357/835/2/224 Astropy Collaboration. 2013, Astronomy & Astrophysics, 558, A33, doi: 10.1051/0004-6361/201322068 Astropy Collaboration. 2018, The Astronomical Journal, 156, 123, doi: 10.3847/1538-3881/aabc4f Astropy Collaboration, Price-Whelan, A. M., ...

    work page doi:10.3847/1538-4357/835/2/224 2017

  8. [8]

    2020, AJ, 160, 18, doi: 10.3847/1538-3881/ab91bf

    Auge, C., Huber, D., Heinze, A., et al. 2020, AJ, 160, 18, doi: 10.3847/1538-3881/ab91bf

    work page doi:10.3847/1538-3881/ab91bf 2020

  9. [9]

    Barnes, S. A. 2007, ApJ, 669, 1167, doi: 10.1086/519295

    work page doi:10.1086/519295 2007

  10. [10]

    2018, ARA&A, 56, 83, doi: 10.1146/annurev-astro-081817-051839

    Bastian, N., & Lardo, C. 2018, ARA&A, 56, 83, doi: 10.1146/annurev-astro-081817-051839

    work page doi:10.1146/annurev-astro-081817-051839 2018

  11. [11]

    L., Werner, S., Mitschang, A

    Beaton, R. L., Werner, S., Mitschang, A. W., et al. 2022, Research Notes of the American Astronomical Society, 6, 125, doi: 10.3847/2515-5172/ac7808

    work page doi:10.3847/2515-5172/ac7808 2022

  12. [12]

    C., Kazantzidis, S., Weinberg, D

    Bird, J. C., Kazantzidis, S., Weinberg, D. H., et al. 2013, ApJ, 773, 43, doi: 10.1088/0004-637X/773/1/43 15

    work page doi:10.1088/0004-637x/773/1/43 2013

  13. [13]

    2026, A&A, 708, A215, doi: 10.1051/0004-6361/202558436

    Boin, T., Casamiquela, L., Haywood, M., et al. 2026, A&A, 708, A215, doi: 10.1051/0004-6361/202558436

    work page doi:10.1051/0004-6361/202558436 2026

  14. [14]

    S., & Vaughan, A

    Bowen, I. S., & Vaughan, A. H., J. 1973, ApOpt, 12, 1430, doi: 10.1364/AO.12.001430

    work page doi:10.1364/ao.12.001430 1973

  15. [15]

    , volume =

    Bressan, A., Marigo, P., Girardi, L., et al. 2012, MNRAS, 427, 127, doi: 10.1111/j.1365-2966.2012.21948.x

    work page doi:10.1111/j.1365-2966.2012.21948.x 2012

  16. [16]

    2021, Monthly Notices of the Royal Astronomical Society, 506, 150, doi: 10.1093/mnras/stab1242

    Buder, S., Sharma, S., Kos, J., et al. 2021, Monthly Notices of the Royal Astronomical Society, 506, 150, doi: 10.1093/mnras/stab1242

    work page doi:10.1093/mnras/stab1242 2021

  17. [17]

    E., et al

    Buder, S., Kos, J., Wang, X. E., et al. 2025, PASA, 42, e051, doi: 10.1017/pasa.2025.26

    work page doi:10.1017/pasa.2025.26 2025

  18. [18]

    Bufanda, E., Tayar, J., Huber, D., Hasselquist, S., & Lane, R. R. 2023, ApJ, 959, 123, doi: 10.3847/1538-4357/acf9a5

    work page doi:10.3847/1538-4357/acf9a5 2023

  19. [19]

    2020, A&A, 640, A1, doi: 10.1051/0004-6361/202038192

    Cantat-Gaudin, T., Anders, F., Castro-Ginard, A., et al. 2020, A&A, 640, A1, doi: 10.1051/0004-6361/202038192

    work page doi:10.1051/0004-6361/202038192 2020

  20. [20]

    2021, A&A, 654, A151, doi: 10.1051/0004-6361/202141779

    Casamiquela, L., Castro-Ginard, A., Anders, F., & Soubiran, C. 2021, A&A, 654, A151, doi: 10.1051/0004-6361/202141779

    work page doi:10.1051/0004-6361/202141779 2021

  21. [21]

    2001, ApJ, 554, 1044, doi: 10.1086/321427

    Chiappini, C., Matteucci, F., & Romano, D. 2001, ApJ, 554, 1044, doi: 10.1086/321427

    work page doi:10.1086/321427 2001

  22. [22]

    MESA Isochrones and Stellar Tracks (MIST). I: Solar-Scaled Models

    Choi, J., Dotter, A., Conroy, C., et al. 2016, ApJ, 823, 102, doi: 10.3847/0004-637X/823/2/102

    work page internal anchor Pith review doi:10.3847/0004-637x/823/2/102 2016

  23. [23]

    2018, ApJ, 860, 131, doi: 10.3847/1538-4357/aac435 Ciuc˘a, I., Kawata, D., Ting, Y.-S., et al

    Choi, J., Dotter, A., Conroy, C., & Ting, Y.-S. 2018, ApJ, 860, 131, doi: 10.3847/1538-4357/aac435 Ciuc˘a, I., Kawata, D., Ting, Y.-S., et al. 2024, MNRAS, 528, L122, doi: 10.1093/mnrasl/slad033

    work page doi:10.3847/1538-4357/aac435 2018

  24. [24]

    Cohen, J. G. 1978, ApJ, 223, 487, doi: 10.1086/156284

    work page doi:10.1086/156284 1978

  25. [25]

    2012, ApJ, 757, 190, doi: 10.1088/0004-637X/757/2/190

    Corsaro, E., Stello, D., Huber, D., et al. 2012, ApJ, 757, 190, doi: 10.1088/0004-637X/757/2/190

    work page doi:10.1088/0004-637x/757/2/190 2012

  26. [26]

    Research in Astronomy and Astrophysics , year = 2012, month = sep, volume =

    Cui, X.-Q., Zhao, Y.-H., Chu, Y.-Q., et al. 2012, Research in Astronomy and Astrophysics, 12, 1197, doi: 10.1088/1674-4527/12/9/003

    work page doi:10.1088/1674-4527/12/9/003 2012

  27. [27]

    L., Ag¨ ueros, M

    Curtis, J. L., Ag¨ ueros, M. A., Matt, S. P., et al. 2020, ApJ, 904, 140, doi: 10.3847/1538-4357/abbf58

    work page doi:10.3847/1538-4357/abbf58 2020

  28. [28]

    M., Badenes, C., Tayar, J., et al

    Daher, C. M., Badenes, C., Tayar, J., et al. 2022, MNRAS, 512, 2051, doi: 10.1093/mnras/stac590 de Jong, R. S., Agertz, O., Berbel, A. A., et al. 2019, The Messenger, 175, 3, doi: 10.18727/0722-6691/5117

    work page doi:10.1093/mnras/stac590 2022

  29. [29]

    El-Badry, K., Rix, H.-W., & Heintz, T. M. 2021, MNRAS, 506, 2269, doi: 10.1093/mnras/stab323

    work page doi:10.1093/mnras/stab323 2021

  30. [30]

    A., et al

    Elsworth, Y., Hekker, S., Johnson, J. A., et al. 2019, MNRAS, 489, 4641, doi: 10.1093/mnras/stz2356

    work page doi:10.1093/mnras/stz2356 2019

  31. [31]

    K., Feltzing, S., Sahlholdt, C., & Casagrande, L

    Feuillet, D. K., Feltzing, S., Sahlholdt, C., & Casagrande, L. 2020, arXiv e-prints, arXiv:2003.11039. https://arxiv.org/abs/2003.11039

    arXiv 2020

  32. [32]

    J., Hogg, D

    Frazer, P., Griffith, E. J., Hogg, D. W., Sinha, A., & Tayar, J. 2025, arXiv e-prints, arXiv:2510.26927, doi: 10.48550/arXiv.2510.26927

    work page doi:10.48550/arxiv.2510.26927 2025

  33. [33]

    2002, ARA&A, 40, 487, doi: 10.1146/annurev.astro.40.060401.093840

    Freeman, K., & Bland-Hawthorn, J. 2002, ARA&A, 40, 487, doi: 10.1146/annurev.astro.40.060401.093840

    work page doi:10.1146/annurev.astro.40.060401.093840 2002

  34. [34]

    M., Thompson, B., Jackson, K

    Frinchaboy, P. M., Thompson, B., Jackson, K. M., et al. 2013, ApJL, 777, L1, doi: 10.1088/2041-8205/777/1/L1 Gaia Collaboration, Brown, A. G. A., Vallenari, A., et al. 2018, A&A, 616, A1, doi: 10.1051/0004-6361/201833051 Gaia Collaboration, Vallenari, A., Brown, A. G. A., et al. 2023, A&A, 674, A1, doi: 10.1051/0004-6361/202243940 Garc´ıa P´erez, A. E., A...

    work page doi:10.1088/2041-8205/777/1/l1 2013

  35. [35]

    J., Becker, I., & Cargile, P

    Garraffo, C., Protopapas, P., Drake, J. J., Becker, I., & Cargile, P. 2021, arXiv e-prints, arXiv:2106.07655. https://arxiv.org/abs/2106.07655

    arXiv 2021

  36. [36]

    , keywords =

    Girardi, L. 1999, MNRAS, 308, 818, doi: 10.1046/j.1365-8711.1999.02746.x

    work page doi:10.1046/j.1365-8711.1999.02746.x 1999

  37. [37]

    H., et al

    Godoy-Rivera, D., Tayar, J., Pinsonneault, M. H., et al. 2021, arXiv e-prints, arXiv:2104.07679. https://arxiv.org/abs/2104.07679

    arXiv 2021

  38. [38]

    G., Sneden, C., Carretta, E., & Bragaglia, A

    Gratton, R. G., Sneden, C., Carretta, E., & Bragaglia, A. 2000, A&A, 354, 169

    2000

  39. [39]

    H., Johnson, J

    Griffith, E., Weinberg, D. H., Johnson, J. A., et al. 2021, ApJ, 909, 77, doi: 10.3847/1538-4357/abd6be

    work page doi:10.3847/1538-4357/abd6be 2021

  40. [40]

    2006, Mem

    Grundahl, F., Kjeldsen, H., Frandsen, S., et al. 2006, Mem. Soc. Astron. Italiana, 77, 458

    2006

  41. [41]

    doi:10.1086/500975 , eprint =

    Gunn, J. E., Siegmund, W. A., Mannery, E. J., et al. 2006, AJ, 131, 2332, doi: 10.1086/500975

    work page doi:10.1086/500975 2006

  42. [42]

    A., Shetrone, M., et al

    Hasselquist, S., Holtzman, J. A., Shetrone, M., et al. 2019, ApJ, 871, 181, doi: 10.3847/1538-4357/aaf859

    work page doi:10.3847/1538-4357/aaf859 2019

  43. [43]

    R., Bovy, J., Holtzman, J

    Hayden, M. R., Bovy, J., Holtzman, J. A., et al. 2015, ApJ, 808, 132, doi: 10.1088/0004-637X/808/2/132

    work page doi:10.1088/0004-637x/808/2/132 2015

  44. [44]

    R., Sharma, S., Bland-Hawthorn, J., et al

    Hayden, M. R., Sharma, S., Bland-Hawthorn, J., et al. 2022, MNRAS, 517, 5325, doi: 10.1093/mnras/stac2787

    work page doi:10.1093/mnras/stac2787 2022

  45. [45]

    H., et al

    Helmi, A., Babusiaux, C., Koppelman, H. H., et al. 2018, Nature, 563, 85, doi: 10.1038/s41586-018-0625-x

    work page doi:10.1038/s41586-018-0625-x 2018

  46. [46]

    2025, ApJS, 280, 14, doi: 10.3847/1538-4365/ade2e9

    Hey, D., Tonry, J., Shappee, B., & Huber, D. 2025, ApJS, 280, 14, doi: 10.3847/1538-4365/ade2e9

    work page doi:10.3847/1538-4365/ade2e9 2025

  47. [47]

    R., Huber, D., Shappee, B

    Hey, D. R., Huber, D., Shappee, B. J., et al. 2023, AJ, 166, 249, doi: 10.3847/1538-3881/ad01bf

    work page doi:10.3847/1538-3881/ad01bf 2023

  48. [48]

    2024, ApJ, 973, 154, doi: 10.3847/1538-4357/ad6320

    Hon, M., Li, Y., & Ong, J. 2024, ApJ, 973, 154, doi: 10.3847/1538-4357/ad6320

    work page doi:10.3847/1538-4357/ad6320 2024

  49. [49]

    S., et al

    Hon, M., Huber, D., Kuszlewicz, J. S., et al. 2021, ApJ, 919, 131, doi: 10.3847/1538-4357/ac14b1

    work page doi:10.3847/1538-4357/ac14b1 2021

  50. [50]

    Horta, D., & Schiavon, R. P. 2024, Ap&SS, 369, 107, doi: 10.1007/s10509-024-04370-y

    work page doi:10.1007/s10509-024-04370-y 2024

  51. [51]

    W., Kalup, C., Stello, D., & De Silva, G

    Howell, M., Campbell, S. W., Kalup, C., Stello, D., & De Silva, G. M. 2025, MNRAS, 536, 1389, doi: 10.1093/mnras/stae2686

    work page doi:10.1093/mnras/stae2686 2025

  52. [52]

    W., Stello, D., & De Silva, G

    Howell, M., Campbell, S. W., Stello, D., & De Silva, G. M. 2022, MNRAS, 515, 3184, doi: 10.1093/mnras/stac1918

    work page doi:10.1093/mnras/stac1918 2022

  53. [53]

    W., Stello, D., & De Silva, G

    Howell, M., Campbell, S. W., Stello, D., & De Silva, G. M. 2024, MNRAS, 527, 7974, doi: 10.1093/mnras/stad3565

    work page doi:10.1093/mnras/stad3565 2024

  54. [54]

    TESS Asteroseismology of Red Giants in the Old Metal-Rich Open Clusters NGC 188 & NGC 6791

    Howell, M., Johnson, J. A., Pinsonneault, M. H., et al. 2026, arXiv e-prints, arXiv:2604.27828, doi: 10.48550/arXiv.2604.27828 16

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2604.27828 2026

  55. [55]

    L., & Reffert, S

    Hunt, E. L., & Reffert, S. 2023, Astronomy & Astrophysics, 673, A114, doi: 10.1051/0004-6361/202346285

    work page doi:10.1051/0004-6361/202346285 2023

  56. [56]

    Hunter, J. D. 2007, Computing in Science and Engineering, 9, 90, doi: 10.1109/MCSE.2007.55

    work page doi:10.1109/mcse.2007.55 2007

  57. [57]

    , keywords =

    Jin, S., Trager, S. C., Dalton, G. B., et al. 2024, MNRAS, 530, 2688, doi: 10.1093/mnras/stad557

    work page doi:10.1093/mnras/stad557 2024

  58. [58]

    W., Weinberg, D

    Johnson, J. W., Weinberg, D. H., Vincenzo, F., et al. 2021, MNRAS, 508, 4484, doi: 10.1093/mnras/stab2718

    work page doi:10.1093/mnras/stab2718 2021

  59. [59]

    W., Weinberg, D

    Johnson, J. W., Weinberg, D. H., Blanc, G. A., et al. 2025, ApJ, 988, 8, doi: 10.3847/1538-4357/addbe5

    work page doi:10.3847/1538-4357/addbe5 2025

  60. [60]

    2018, ApJ, 864, 99, doi: 10.3847/1538-4357/aad464

    Joyce, M., & Chaboyer, B. 2018, ApJ, 864, 99, doi: 10.3847/1538-4357/aad464

    work page doi:10.3847/1538-4357/aad464 2018

  61. [61]

    I., Marchetti, T., et al

    Joyce, M., Johnson, C. I., Marchetti, T., et al. 2023, ApJ, 946, 28, doi: 10.3847/1538-4357/acb692

    work page doi:10.3847/1538-4357/acb692 2023

  62. [62]

    P., & Ba, J

    Kingma, D. P., & Ba, J. 2017, https://arxiv.org/abs/1412.6980

    Pith/arXiv arXiv 2017

  63. [63]

    N., Guhathakurta, P., Zhang, A

    Kirby, E. N., Guhathakurta, P., Zhang, A. J., et al. 2016, ApJ, 819, 135, doi: 10.3847/0004-637X/819/2/135

    work page doi:10.3847/0004-637x/819/2/135 2016

  64. [64]

    Kjeldsen, H., & Bedding, T. R. 1995, A&A, 293, 87

    1995

  65. [65]

    A., Rix, H.-W., Aerts, C., et al

    Kollmeier, J. A., Rix, H.-W., Aerts, C., et al. 2026, AJ, 171, 52, doi: 10.3847/1538-3881/ae0576

    work page doi:10.3847/1538-3881/ae0576 2026

  66. [66]

    W., & Bovy, J

    Leung, H. W., & Bovy, J. 2019, MNRAS, 483, 3255, doi: 10.1093/mnras/sty3217

    work page doi:10.1093/mnras/sty3217 2019

  67. [67]

    W., Bovy, J., Mackereth, J

    Leung, H. W., Bovy, J., Mackereth, J. T., & Miglio, A. 2023, MNRAS, 522, 4577, doi: 10.1093/mnras/stad1272

    work page doi:10.1093/mnras/stad1272 2023

  68. [68]

    R., Murphy, S

    Li, Y., Bedding, T. R., Murphy, S. J., et al. 2022, Nature Astronomy, 6, 673, doi: 10.1038/s41550-022-01648-5

    work page doi:10.1038/s41550-022-01648-5 2022

  69. [69]

    R., Huber, D., et al

    Li, Y., Bedding, T. R., Huber, D., et al. 2024, ApJ, 974, 77, doi: 10.3847/1538-4357/ad6c3e

    work page doi:10.3847/1538-4357/ad6c3e 2024

  70. [70]

    Li, Y., Huber, D., Ong, J. M. J., et al. 2025, ApJ, 984, 125, doi: 10.3847/1538-4357/adc737

    work page doi:10.3847/1538-4357/adc737 2025

  71. [71]

    LSST Science Book, Version 2.0

    Lian, J., Zasowski, G., Hasselquist, S., et al. 2022, MNRAS, 511, 5639, doi: 10.1093/mnras/stac479 LSST Science Collaboration, Abell, P. A., Allison, J., et al. 2009, arXiv e-prints, arXiv:0912.0201, doi: 10.48550/arXiv.0912.0201

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1093/mnras/stac479 2022

  72. [72]

    H., Ting, Y.-S., et al

    Lu, Y., Pinsonneault, M. H., Ting, Y.-S., et al. 2026, AJ, 171, 38, doi: 10.3847/1538-3881/ae196b

    work page doi:10.3847/1538-3881/ae196b 2026

  73. [73]

    L., Angus, R., Curtis, J

    Lu, Y. L., Angus, R., Curtis, J. L., David, T. J., & Kiman, R. 2021, AJ, 161, 189, doi: 10.3847/1538-3881/abe4d6

    work page doi:10.3847/1538-3881/abe4d6 2021

  74. [74]

    T., Bovy, J., Leung, H

    Mackereth, J. T., Bovy, J., Leung, H. W., et al. 2019, MNRAS, 489, 176, doi: 10.1093/mnras/stz1521

    work page doi:10.1093/mnras/stz1521 2019

  75. [75]

    2025, arXiv e-prints, arXiv:2512.20923, doi: 10.48550/arXiv.2512.20923

    Mankowski, C., Tayar, J., & Martin, C. 2025, arXiv e-prints, arXiv:2512.20923, doi: 10.48550/arXiv.2512.20923

    work page doi:10.48550/arxiv.2512.20923 2025

  76. [76]

    2016, MNRAS, 456, 3655, doi: 10.1093/mnras/stv2830 M´esz´aros, S., Jofr´e, P., Johnson, J

    Martig, M., Fouesneau, M., Rix, H.-W., et al. 2016, MNRAS, 456, 3655, doi: 10.1093/mnras/stv2830 M´esz´aros, S., Jofr´e, P., Johnson, J. A., et al. 2025, AJ, 170, 96, doi: 10.3847/1538-3881/ade4b9

    work page doi:10.1093/mnras/stv2830 2016

  77. [77]

    T., et al

    Miglio, A., Chiappini, C., Mackereth, J. T., et al. 2021a, A&A, 645, A85, doi: 10.1051/0004-6361/202038307

    work page doi:10.1051/0004-6361/202038307

  78. [78]

    2021b, Experimental Astronomy, 51, 963, doi: 10.1007/s10686-021-09711-1

    Miglio, A., Girardi, L., Grundahl, F., et al. 2021b, Experimental Astronomy, 51, 963, doi: 10.1007/s10686-021-09711-1

    work page doi:10.1007/s10686-021-09711-1

  79. [79]

    I., et al

    Miller, N., Joyce, M., Johnson, C. I., et al. 2025, arXiv e-prints, arXiv:2512.08090, doi: 10.48550/arXiv.2512.08090

    work page doi:10.48550/arxiv.2512.08090 2025

  80. [80]

    M., Tayar, J., & Claytor, Z

    Morales, L. M., Tayar, J., & Claytor, Z. R. 2025, arXiv e-prints, arXiv:2504.17600, doi: 10.48550/arXiv.2504.17600

    work page doi:10.48550/arxiv.2504.17600 2025

Showing first 80 references.