pith. sign in

arxiv: 2606.21035 · v1 · pith:RLKVWF7Unew · submitted 2026-06-19 · 🌌 astro-ph.IM

Quality Assessment of Spectroscopic Data Reduction Pipelines Using Artificial Intelligence: Scrutinizing Data Release 2 from the DESI Survey

V. Torres-Gomez , J. Suarez-Perez , J. E. Forero-Romero , S. Bailey , A. Kremin , B. Dey , R. P. Nathan , S. Panda
show 40 more authors
J. Aguilar S. Ahlen D. Bianchi D. Brooks T. Claybaugh A. de la Macorra P. Doel S. Ferraro A. Font-Ribera E. Gaztanaga S. Gontcho A Gontcho G. Gutierrez K. Honscheid C. Howlett R. Joyce S. Juneau D. Kirkby O. Lahav M. Landriau L. Le Guillou M. E. Levi M. Manera A. Meisner R. Miquel J. Moustakas S. Nadathur N. Palanque-Delabrouille W. J. Percival F. Prada I. Perez-Rafols G. Rossi L. Samushia E. Sanchez D. Schlegel H. Seo R. Sharples J. Silber G. Tarle B. A. Weaver H. Zou
This is my paper

Pith reviewed 2026-06-26 13:43 UTC · model grok-4.3

classification 🌌 astro-ph.IM
keywords DESIspectroscopic quality assessmentoutlier detectionUMAPFriends-of-Friends clusteringdata reduction pipelineunsupervised learningDESI DR2
0
0 comments X

The pith

An unsupervised clustering method flags reduction artifacts in 67 percent of DESI outlier spectra that standard pipelines leave unflagged.

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

The paper introduces an unsupervised pipeline that uses dimensionality reduction and clustering to isolate anomalous spectra in massive datasets without any labeled training examples. It processes each of the 14,199 tiles from DESI Data Release 2 separately, identifying over one million outlier candidates from 58 million total spectra. Visual review of 391 candidates shows that two-thirds display clear spectral anomalies tied to known reduction and calibration problems, while the standard pipeline's quality flags catch only 4 percent of them. This establishes the method as a practical additional check that catches issues missed by existing diagnostics at survey scale.

Core claim

The pipeline applies UMAP dimensionality reduction and Friends-of-Friends clustering independently to each tile, separating a dense core of typical spectra from small isolated groups and singletons that total 1,095,816 candidates. Inspection of 391 sampled outliers from the main survey programs finds that 66.8 percent show identifiable reduction and calibration effects, while only 4.1 percent carry non-zero quality flags from the standard pipeline. Extrapolating to the full main-survey catalog yields an estimate of roughly 218,000 candidates free of such artifacts.

What carries the argument

UMAP dimensionality reduction combined with Friends-of-Friends clustering applied independently per tile to separate typical spectra from isolated outliers.

If this is right

  • The method supplies a complementary quality-assessment layer that recovers a substantial population of problematic spectra missed by standard diagnostics.
  • Roughly 218,000 candidate outliers appear free of reduction artifacts and may represent genuine atypical spectra.
  • The approach is scalable and reproducible across successive data releases for ongoing monitoring.
  • Mean outlier fractions are 0.76 percent in dark programs and 2.36 percent in bright programs.

Where Pith is reading between the lines

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

  • The same per-tile clustering workflow could be applied to data from other large spectroscopic surveys to surface hidden reduction issues.
  • The estimated 218,000 artifact-free outliers merit targeted follow-up observations to determine whether they correspond to rare astronomical objects.
  • Adjusting clustering parameters with feedback from more inspections could sharpen the separation between artifacts and valid but unusual spectra.

Load-bearing premise

The selected UMAP and Friends-of-Friends parameters per tile isolate reduction artifacts rather than genuine rare spectra, and the 391 inspected candidates represent the full set of over one million outliers.

What would settle it

A larger visual inspection of several thousand randomly selected candidates that finds the fraction with identifiable reduction anomalies falling below 50 percent would undermine the claim that the method reliably recovers missed problematic spectra.

Figures

Figures reproduced from arXiv: 2606.21035 by A. de la Macorra, A. Font-Ribera, A. Kremin, A. Meisner, B. A. Weaver, B. Dey, C. Howlett, D. Bianchi, D. Brooks, D. Kirkby, D. Schlegel, E. Gaztanaga, E. Sanchez, F. Prada, G. Gutierrez, G. Rossi, G. Tarle, H. Seo, H. Zou, I. Perez-Rafols, J. Aguilar, J. E. Forero-Romero, J. Moustakas, J. Silber, J. Suarez-Perez, K. Honscheid, L. Le Guillou, L. Samushia, M. E. Levi, M. Landriau, M. Manera, N. Palanque-Delabrouille, O. Lahav, P. Doel, R. Joyce, R. Miquel, R. P. Nathan, R. Sharples, S. Ahlen, S. Bailey, S. Ferraro, S. Gontcho A Gontcho, S. Juneau, S. Nadathur, S. Panda, T. Claybaugh, V. Torres-Gomez, W. J. Percival.

Figure 1
Figure 1. Figure 1: — DESI focal plane. Mean fiber positions are aggregated over the tiles analyzed in this work. Blue points show all 5,000 spectroscopic fiber positions, while orange points show the subset of fiber positions that were assigned to science targets in at least one of those tiles. Petal identifiers (0–9) are annotated; each petal has 500 fibers feeding one bench spectrograph. per exposure is below 5,000. Fields… view at source ↗
Figure 2
Figure 2. Figure 2: — UMAP embedding for TILEID 8643, containing 4,105 spectra after quality and target-type selection, and colored by redrock spectroscopic class (STAR, GALAXY, QSO). Galaxies dominate the main embedding, while stars and quasars appear with substantial overlap with the galaxy distribution. spectral classes. For the FoF stage, a linking length of ℓ = 0.15 and a minimum group size of Nmin = 5 were adopted; thes… view at source ↗
Figure 3
Figure 3. Figure 3: — Same embedding as [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: — UMAP embedding for TILEID 8643. Gray points show all spectra; black crosses mark candidates (singletons and groups with |C| < Nmin). spectra, together with a small number of compact groups and isolated singletons in the low-density borders. These peripheral points are the candidate outliers [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: — Outlier fraction fout per tile across DR2, separated by observing program (Dark, Bright, and Backup), shown with fixed-width bins and a logarithmic count axis. Normalizing by the number of spectra reduces sample-size effects. Most tiles have candidate fractions of a few percent or less; tiles with fout > 0.2 are rare and occur primarily in the Backup program [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: — Distribution of candidate outlier fraction per science fiber across DR2, separated by observing program. Most fibers show fractions at the percent level, and a small high-fraction tail is present [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: — Focal plane map of outlier fraction per fiber. Points mark science fibers at their mean focal plane coordinates; color encodes outlier candidate fraction, and petal identifiers (0–9) are annotated. The distribution is largely uniform, with localized re￾gions of elevated fractions. plane described in Section 2.1. The baseline level of the candidate fraction increases from the dark program to the bright an… view at source ↗
Figure 8
Figure 8. Figure 8: — Candidate outlier fraction per PETALID across DR2, separated by observing program. This normalization accounts for differences in petal contribution across tiles and observing programs. All petals contribute non-zero fractions, with substantial petal-to-petal variation [PITH_FULL_IMAGE:figures/full_fig_p010_8.png] view at source ↗
Figure 9
Figure 9. Figure 9: — Candidate outlier fractions per fiber as a function of FIBERID, separated by observing program. The ∼500-fiber periodicity reflects the petal/spectrograph segmentation of the focal plane. Candidate fractions increase toward the edges of each 500-fiber spectrograph block, consistent with small edge-dependent differences in spectral resolution, focus, throughput, or noise properties. the candidate set. The… view at source ↗
Figure 10
Figure 10. Figure 10: — [PITH_FULL_IMAGE:figures/full_fig_p010_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: — Candidate fraction by observing program on a loga￾rithmic scale. All programs contribute, spanning a range of candi￾date fractions. The variation indicates that the outlier rate depends on observing conditions and target populations [PITH_FULL_IMAGE:figures/full_fig_p011_11.png] view at source ↗
Figure 12
Figure 12. Figure 12: — Histogram of wall-clock time per tile, ttile, in seconds across DR2. Runtimes concentrate around a typical value, with a minority of outlying tiles. The narrow runtime distribution in￾dicates that the implementation has a predictable computational cost for routine tile-level processing. to produce more candidates, though the scatter is large. The discrete vertical bands in Nspec reflect common tile conf… view at source ↗
Figure 14
Figure 14. Figure 14: — Galaxy targeted as an LRG and classified by redrock as SPECTYPE=GALAXY, showing arm-join flux mismatches near 5,800 Å (B–R boundary) and 7,600 Å (R–Z boundary). The continuum shifts abruptly at each arm boundary while the redrock fit (black) remains smooth (TARGETID 39628374682901440, TILEID 3473, FIBERID 470) [PITH_FULL_IMAGE:figures/full_fig_p013_14.png] view at source ↗
Figure 15
Figure 15. Figure 15: — Galaxy targeted as a BGS object and classified by redrock as SPECTYPE=GALAXY, showing spurious Z-arm emission. A strong, narrow peak appears in the Z arm with no associated redrock emission line at the fitted redshift, while the B and R arms follow the pipeline fit (black) closely. The feature is consistent with sky-subtraction residuals (TARGETID 39633232366404175, TILEID 22175, FIBERID 3881) [PITH_FU… view at source ↗
Figure 16
Figure 16. Figure 16: — Galaxy targeted as an LRG and classified by redrock as SPECTYPE=GALAXY, showing a depressed blue continuum. The B arm (blue) is depressed below zero over a broad wavelength interval, while the R and Z arms (red) remain positive. The black curve is the redrock best-fit model (TARGETID 39633251953804852, TILEID 25300, FIBERID 4393). tra; it produces a ranked watch list that concentrates human review where… view at source ↗
read the original abstract

Large spectroscopic surveys now collect data at a scale that makes traditional visual inspection impractical. We present an unsupervised pipeline for spectroscopic quality assessment that requires no labeled training data. The method combines Uniform Manifold Approximation and Projection for dimensionality reduction with Friends-of-Friends clustering to isolate anomalous spectra for targeted review. We apply this pipeline to 58,291,334 spectra across 14,199 tiles from DESI Data Release 2, processing each tile independently to produce a tile-level outlier catalog. In each tile, the pipeline separates a dense core of typical spectra from small, isolated components and singletons, yielding a total of 1,095,816 outlier candidates. The mean tile-level outlier fraction is about 1.96 percent overall, with values of 0.76 percent and 2.36 percent for the dark and bright main-survey programs, respectively. From the visual inspection of 391 outlier candidates from the dark and bright programs of the main survey, we find that 66.8 percent exhibit identifiable spectral anomalies consistent with known reduction and calibration effects. By contrast, only 4.1 percent carry a non-zero quality flag from the standard reduction pipeline. This shows that the method provides a complementary quality-assessment layer to existing pipeline diagnostics and recovers a substantial population of problematic spectra that standard diagnostics miss. Extrapolating to the main-survey catalog, we estimate that approximately 218,000 candidate outliers are free of identifiable reduction artifacts and may correspond to genuine atypical spectra in the context of DESI. The pipeline is scalable, reproducible, and directly comparable across successive data releases, making it a practical quality-assurance monitor for DESI and future multi-object spectroscopic surveys.

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 / 2 minor

Summary. The paper presents an unsupervised pipeline combining UMAP dimensionality reduction and Friends-of-Friends clustering to detect anomalous spectra in DESI DR2 without labeled training data. Applied tile-by-tile to 58,291,334 spectra, it identifies 1,095,816 outlier candidates (mean fraction ~1.96%). Visual inspection of 391 candidates from dark/bright main-survey programs finds 66.8% with identifiable reduction/calibration anomalies, compared to only 4.1% flagged by the standard pipeline. The work claims complementarity to existing diagnostics and extrapolates to ~218,000 genuine atypical spectra, positioning the method as a scalable QA monitor for DESI and future surveys.

Significance. If the central results hold, the approach offers a practical, reproducible, label-free tool for quality assessment at the scale of modern spectroscopic surveys, directly addressing the impracticality of full visual inspection. Strengths include the tile-independent processing (avoiding global parameter biases), the large application to 14,199 tiles, and the explicit comparison to standard quality flags. The extrapolation to 218k genuine outliers provides a falsifiable prediction for targeted follow-up. However, the significance is tempered by the small inspected sample size and lack of reported sampling/parameter details, which limit confidence in the complementarity claim.

major comments (1)
  1. [Abstract / Results] Abstract and Results (visual inspection paragraph): The claim that 66.8% of inspected candidates show reduction artifacts (vs. 4.1% standard flags) is load-bearing for the complementarity conclusion and the 218k extrapolation. The manuscript provides no information on how the 391 candidates were selected from the 1,095,816 (random, score-stratified, or otherwise), nor on inter-rater reliability or exact UMAP/FoF hyperparameters per tile. Without this, the inspected fraction cannot be shown to be representative, undermining generalization to the full set.
minor comments (2)
  1. [Abstract] The reported mean outlier fractions (0.76% dark, 2.36% bright) would benefit from accompanying standard deviations or tile-to-tile histograms to convey variability.
  2. [Methods] Notation for the outlier catalog (e.g., how singletons vs. small clusters are defined) should be clarified with a brief equation or pseudocode in the methods.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for their thoughtful review and for identifying a point that requires clarification to strengthen the manuscript. We address the major comment below and will revise the paper accordingly.

read point-by-point responses

  1. Referee: [Abstract / Results] Abstract and Results (visual inspection paragraph): The claim that 66.8% of inspected candidates show reduction artifacts (vs. 4.1% standard flags) is load-bearing for the complementarity conclusion and the 218k extrapolation. The manuscript provides no information on how the 391 candidates were selected from the 1,095,816 (random, score-stratified, or otherwise), nor on inter-rater reliability or exact UMAP/FoF hyperparameters per tile. Without this, the inspected fraction cannot be shown to be representative, undermining generalization to the full set.

    Authors: We agree that the manuscript does not provide sufficient detail on the selection of the 391 inspected candidates, the precise UMAP and FoF hyperparameters used on a per-tile basis, or inter-rater reliability for the visual inspection. These omissions limit the ability of readers to evaluate the representativeness of the inspected sample. In the revised manuscript we will add a new subsection (or expand the existing Methods/Results section) that explicitly describes the sampling procedure used to choose the 391 spectra, the hyperparameter values (or ranges) applied tile-by-tile, and any steps taken to assess consistency in the visual classifications. This addition will directly address the concern and support the generalization to the full outlier catalog. revision: yes

Circularity Check

0 steps flagged

No circularity: unsupervised data-driven method with independent validation

full rationale

The derivation applies standard UMAP dimensionality reduction and Friends-of-Friends clustering independently per tile to raw spectra, with no labeled data, no parameter fitting to target quality flags, and no self-referential definitions. The 66.8% anomaly rate comes from separate visual inspection of a sample, not from any equation or prior result that reduces to the method's own outputs. No self-citations, uniqueness theorems, or ansatzes are invoked as load-bearing steps. The pipeline is self-contained against external benchmarks and does not rename known results or smuggle assumptions via citation.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only review reveals no explicit free parameters, axioms, or invented entities beyond reliance on standard UMAP and FoF algorithms; any clustering parameters are implicit but unspecified.

pith-pipeline@v0.9.1-grok · 6102 in / 1259 out tokens · 41769 ms · 2026-06-26T13:43:55.948154+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

57 extracted references · 50 canonical work pages · 4 internal anchors

  1. [1]

    2025, Journal of Cosmology and Astroparticle Physics, 2025, 028, doi: 10.1088/1475-7516/2025/07/028

    Adame, A., Aguilar, J., Ahlen, S., et al. 2025, Journal of Cosmology and Astroparticle Physics, 2025, 028, doi: 10.1088/1475-7516/2025/07/028

    work page doi:10.1088/1475-7516/2025/07/028 2025

  2. [2]

    , year=2024, month=aug, volume=

    Adame, A. G., Aguilar, J., Ahlen, S., et al. 2024, The Astronomical Journal, 168, 58, doi: 10.3847/1538-3881/ad3217 16

    work page doi:10.3847/1538-3881/ad3217 2024

  3. [3]

    M., Davis, T

    Alexander, D. M., Davis, T. M., Chaussidon, E., et al. 2023, The Astronomical Journal, 165, 124, doi: 10.3847/1538-3881/acacfc

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

  4. [4]

    2024, arXiv e-prints, arXiv:2405.12518, doi: 10.48550/arXiv.2405.12518

    Bacon, R., Maineiri, V., Randich, S., et al. 2024, arXiv e-prints, arXiv:2405.12518, doi: 10.48550/arXiv.2405.12518

    work page doi:10.48550/arxiv.2405.12518 2024

  5. [5]

    2016, Monthly Notices of the Royal Astronomical Society, 465, 4530–4555, doi: 10.1093/mnras/stw3021

    Baron, D., & Poznanski, D. 2016, Monthly Notices of the Royal Astronomical Society, 465, 4530–4555, doi: 10.1093/mnras/stw3021

    work page doi:10.1093/mnras/stw3021 2016

  6. [6]

    2025, Journal of Cosmology and Astroparticle Physics, 2025, 130, doi: 10.1088/1475-7516/2025/01/130

    Bault, A., Kirkby, D., Guy, J., et al. 2025, Journal of Cosmology and Astroparticle Physics, 2025, 130, doi: 10.1088/1475-7516/2025/01/130

    work page doi:10.1088/1475-7516/2025/01/130 2025

  7. [7]

    2025, arXiv e-prints, arXiv:2503.07923, doi: 10.48550/arXiv.2503.07923

    Besuner, R., Dey, A., Drlica-Wagner, A., et al. 2025, arXiv e-prints, arXiv:2503.07923, doi: 10.48550/arXiv.2503.07923

    work page doi:10.48550/arxiv.2503.07923 2025

  8. [8]

    Breunig, Hans-Peter Kriegel, Raymond T

    Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. 2000a, SIGMOD Rec., 29, 93–104, doi: 10.1145/335191.335388

    work page doi:10.1145/335191.335388

  9. [9]

    M., Kriegel, H.-P., Ng, R

    Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. 2000b, in Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data, SIGMOD ’00 (New York,

    2000

  10. [10]

    NY, USA: Association for Computing Machinery), 93–104, doi: 10.1145/342009.335388

    work page doi:10.1145/342009.335388

  11. [11]

    2023, The Astronomical Journal, 166, 66, doi: 10.3847/1538-3881/ace35d

    Brodzeller, A., Dawson, K., Bailey, S., et al. 2023, The Astronomical Journal, 166, 66, doi: 10.3847/1538-3881/ace35d

    work page doi:10.3847/1538-3881/ace35d 2023

  12. [12]

    2023, The Astrophysical Journal, 944, 107, doi: 10.3847/1538-4357/acb3c2

    Chaussidon, E., Yèche, C., Palanque-Delabrouille, N., et al. 2023, The Astrophysical Journal, 944, 107, doi: 10.3847/1538-4357/acb3c2

    work page doi:10.3847/1538-4357/acb3c2 2023

  13. [13]

    2020, The Messenger, 180, 10

    Cirasuolo, M., et al. 2020, The Messenger, 180, 10

    2020

  14. [14]

    J., & Pearson, E

    Clopper, C. J., & Pearson, E. S. 1934, Biometrika, 26, 404, doi: 10.1093/biomet/26.4.404

    work page doi:10.1093/biomet/26.4.404 1934

  15. [15]

    2016, The DESI Experiment Part II: Instrument Design

    Collaboration, D., Aghamousa, A., Aguilar, J., et al. 2016, The DESI Experiment Part II: Instrument Design. https://arxiv.org/abs/1611.00037

    Pith/arXiv arXiv 2016

  16. [16]

    L., Bandi, B., Philipsborn, S., et al

    Cook, T. L., Bandi, B., Philipsborn, S., et al. 2024, Monthly Notices of the Royal Astronomical Society, 535, 2129–2148, doi: 10.1093/mnras/stae2389

    work page doi:10.1093/mnras/stae2389 2024

  17. [17]

    and Koposov, Sergey E

    Cooper, A. P., Koposov, S. E., Allende Prieto, C., et al. 2023, The Astrophysical Journal, 947, 37, doi: 10.3847/1538-4357/acb3c0

    work page doi:10.3847/1538-4357/acb3c0 2023

  18. [18]

    DESI DR2 Results II: Measurements of Baryon Acoustic Oscillations and Cosmological Constraints

    Dalton, G., et al. 2024, Monthly Notices of the Royal Astronomical Society, 530, 2688 de Jong, R. S., Agertz, O., Berbel, A. A., et al. 2019, The Messenger, 175, 3 DESI Collaboration. 2025a, Phys. Rev. D, 112, 083515. https://arxiv.org/abs/2503.14738 —. 2025b, Phys. Rev. D, 112, 083514. https://arxiv.org/abs/2503.14739 DESI Collaboration, Adame, A. G., Ag...

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.1088/1475-7516/2025/07/017 2024

  19. [19]

    , author =

    Dey, A., Schlegel, D. J., Lang, D., et al. 2019, The Astronomical Journal, 157, 168, doi: 10.3847/1538-3881/ab089d

    work page doi:10.3847/1538-3881/ab089d 2019

  20. [20]

    , author =

    Guy, J., Bailey, S., Kremin, A., et al. 2023, The Astronomical Journal, 165, 144, doi: 10.3847/1538-3881/acb212

    work page doi:10.3847/1538-3881/acb212 2023

  21. [21]

    , author =

    Hahn, C., Wilson, M. J., Ruiz-Macias, O., et al. 2023, The Astronomical Journal, 165, 253, doi: 10.3847/1538-3881/accff8

    work page doi:10.3847/1538-3881/accff8 2023

  22. [22]

    P., & Geller, M

    Huchra, J. P., & Geller, M. J. 1982, Astrophysical Journal, 257, 423, doi: 10.1086/160000

    work page doi:10.1086/160000 1982

  23. [23]

    M., et al

    Juneau, S., Canning, R., Alexander, D. M., et al. 2025, The Astronomical Journal, 169, 157, doi: 10.3847/1538-3881/adab0d

    work page doi:10.3847/1538-3881/adab0d 2025

  24. [24]

    SDSS-V: Pioneering Panoptic Spectroscopy

    Kollmeier, J. A., Zasowski, G., Rix, H.-W., et al. 2017, arXiv e-prints, arXiv:1711.03234, doi: 10.48550/arXiv.1711.03234

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1711.03234 2017

  25. [25]

    J., et al

    Krolewski, A., Yu, J., Ross, A. J., et al. 2025, J. Cosmology Astropart. Phys., 2025, 147, doi: 10.1088/1475-7516/2025/01/147

    work page doi:10.1088/1475-7516/2025/01/147 2025

  26. [26]

    2023, The Astrophysical Journal, 943, 68, doi: 10.3847/1538-4357/aca5fa

    Lan, T.-W., Tojeiro, R., Armengaud, E., et al. 2023, The Astrophysical Journal, 943, 68, doi: 10.3847/1538-4357/aca5fa

    work page doi:10.3847/1538-4357/aca5fa 2023

  27. [27]

    The DESI Experiment, a whitepaper for Snowmass 2013

    Levi, M., Bebek, C., Beers, T., et al. 2013, arXiv e-prints, arXiv:1308.0847, doi: 10.48550/arXiv.1308.0847

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.1308.0847 2013

  28. [28]

    2023, ApJ, 956, L6, doi: 10.3847/2041-8213/acfa03

    Liang, Y., Melchior, P., Hahn, C., et al. 2023, ApJ, 956, L6, doi: 10.3847/2041-8213/acfa03

    work page doi:10.3847/2041-8213/acfa03 2023

  29. [29]

    Isolation forest,

    Liu, F. T., Ting, K. M., & Zhou, Z.-H. 2008, in 2008 Eighth IEEE International Conference on Data Mining, 413–422, doi: 10.1109/ICDM.2008.17

    work page doi:10.1109/icdm.2008.17 2008

  30. [30]

    C., Kimball, A

    McGurk, R. C., Kimball, A. E., & Ivezić, Ž. 2010, The Astronomical Journal, 139, 1261, doi: 10.1088/0004-6256/139/3/1261

    work page doi:10.1088/0004-6256/139/3/1261 2010

  31. [31]

    2020, UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction

    McInnes, L., Healy, J., & Melville, J. 2020, UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction. https://arxiv.org/abs/1802.03426

    Pith/arXiv arXiv 2020

  32. [32]

    N., Doel, P., Gutierrez, G., et al

    Miller, T. N., Doel, P., Gutierrez, G., et al. 2023, The Optical Corrector for the Dark Energy Spectroscopic Instrument. https://arxiv.org/abs/2306.06310

    arXiv 2023

  33. [33]

    , author =

    Moustakas, J., Lang, D., Dey, A., et al. 2023, The Astrophysical Journal Supplement Series, 269, 3, doi: 10.3847/1538-4365/acfaa2

    work page doi:10.3847/1538-4365/acfaa2 2023

  34. [34]

    2022, Monthly Notices of the Royal Astronomical Society, 517, 393–419, doi: 10.1093/mnras/stac2582

    Narayan, G. 2022, Monthly Notices of the Royal Astronomical Society, 517, 393–419, doi: 10.1093/mnras/stac2582

    work page doi:10.1093/mnras/stac2582 2022

  35. [35]

    D., Moustakas, J., Bailey, S., et al

    Myers, A. D., Moustakas, J., Bailey, S., et al. 2023, The Astronomical Journal, 165, 50, doi: 10.3847/1538-3881/aca5f9

    work page doi:10.3847/1538-3881/aca5f9 2023

  36. [36]

    P., Lahav, O., et al

    Nicolaou, C., Nathan, R. P., Lahav, O., et al. 2026, Monthly Notices of the Royal Astronomical Society, 547, stag010, doi: 10.1093/mnras/stag010

    work page doi:10.1093/mnras/stag010 2026

  37. [37]

    2014, The Astrophysical Journal, 793, 23, doi: 10.1088/0004-637x/793/1/23

    Nun, I., Pichara, K., Protopapas, P., & Kim, D.-W. 2014, The Astrophysical Journal, 793, 23, doi: 10.1088/0004-637x/793/1/23

    work page doi:10.1088/0004-637x/793/1/23 2014

  38. [38]

    2025, Astronomy & Astrophysics, 703, A242, doi: 10.1051/0004-6361/202556339

    Ortiz, E., & Boquien, M. 2025, Astronomy & Astrophysics, 703, A242, doi: 10.1051/0004-6361/202556339

    work page doi:10.1051/0004-6361/202556339 2025

  39. [39]

    , author =

    Poppett, C., Tyas, L., Aguilar, J., et al. 2024, The Astronomical Journal, 168, 245, doi: 10.3847/1538-3881/ad76a4

    work page doi:10.3847/1538-3881/ad76a4 2024

  40. [40]

    Portillo, S. K. N., Parejko, J. K., Vergara, J. R., & Connolly, A. J. 2020, The Astronomical Journal, 160, 45, doi: 10.3847/1538-3881/ab9644

    work page doi:10.3847/1538-3881/ab9644 2020

  41. [41]

    J., Karim, T., et al

    Raichoor, A., Eisenstein, D. J., Karim, T., et al. 2020, Research Notes of the AAS, 4, 180, doi: 10.3847/2515-5172/abc078

    work page doi:10.3847/2515-5172/abc078 2020

  42. [42]

    , keywords =

    Raichoor, A., Moustakas, J., Newman, J. A., et al. 2023, The Astronomical Journal, 165, 126, doi: 10.3847/1538-3881/acb213

    work page doi:10.3847/1538-3881/acb213 2023

  43. [43]

    S., Bignone, L

    Rosito, M. S., Bignone, L. A., Tissera, P. B., & Pedrosa, S. E. 2023, Astronomy & Astrophysics, 671, A19, doi: 10.1051/0004-6361/202244707

    work page doi:10.1051/0004-6361/202244707 2023

  44. [44]

    J., Beutler, F., Chuang, C.-H., et al

    Ross, A. J., Beutler, F., Chuang, C.-H., et al. 2016, Monthly Notices of the Royal Astronomical Society, 464, 1168–1191, doi: 10.1093/mnras/stw2372

    work page doi:10.1093/mnras/stw2372 2016

  45. [45]

    J., Aguilar, J., Ahlen, S., et al

    Ross, A. J., Aguilar, J., Ahlen, S., et al. 2025, Journal of Cosmology and Astroparticle Physics, 2025, 020, doi: 10.1088/1475-7516/2025/01/020

    work page doi:10.1088/1475-7516/2025/01/020 2025

  46. [46]

    2021, Monthly Notices of the Royal Astronomical Society, 502, 4328, doi: 10.1093/mnras/stab292

    Ruiz-Macias, O., Zarrouk, P., Cole, S., et al. 2021, Monthly Notices of the Royal Astronomical Society, 502, 4328, doi: 10.1093/mnras/stab292

    work page doi:10.1093/mnras/stab292 2021

  47. [47]

    , author =

    Schlafly, E. F., Kirkby, D., Schlegel, D. J., et al. 2023, AJ, 166, 259, doi: 10.3847/1538-3881/ad0832

    work page doi:10.3847/1538-3881/ad0832 2023

  48. [48]

    2023, Monthly Notices of the Royal Astronomical Society, 526, 585, doi: 10.1093/mnras/stad2668

    Sharbaf, Z., Ferreras, I., & Lahav, O. 2023, Monthly Notices of the Royal Astronomical Society, 526, 585, doi: 10.1093/mnras/stad2668

    work page doi:10.1093/mnras/stad2668 2023

  49. [49]

    , author =

    Silber, J. H., Fagrelius, P., Fanning, K., et al. 2022, The Astronomical Journal, 165, 9, doi: 10.3847/1538-3881/ac9ab1 Suárez-Pérez, J. F., & Forero-Romero, J. 2025, in IAU

    work page doi:10.3847/1538-3881/ac9ab1 2022

  50. [50]

    19, IAU Symposium, ed

    Symposium, Vol. 19, IAU Symposium, ed. J. McIver, A. Mahabal, & C. Fluke, 91–94, doi: 10.1017/S1743921322003568 Suárez Pérez, J. F. 2023, PhD thesis, Universidad de los Andes. https://hdl.handle.net/1992/68996 Sánchez-Sáez, P., Lira, H., Martí, L., et al. 2021, The Astronomical Journal, 162, 206, doi: 10.3847/1538-3881/ac1426

    work page doi:10.1017/s1743921322003568 2023

  51. [51]

    2016, Proceedings of the SPIE, 9908, 99081A van der Maaten, L., & Hinton, G

    Takada, M., et al. 2016, Proceedings of the SPIE, 9908, 99081A van der Maaten, L., & Hinton, G. 2008, Journal of Machine Learning Research, 9, 2579

    2016

  52. [52]

    J., Scargle, J

    Way, M. J., Scargle, J. D., Ali, K., & Srivastava, A. N. 2012, Advances in Machine Learning and Data Mining for

    2012

  53. [53]

    W., Connolly, A

    Yip, C. W., Connolly, A. J., Szalay, A. S., et al. 2004a, The Astronomical Journal, 128, 585–609, doi: 10.1086/422429

    work page doi:10.1086/422429

  54. [54]

    W., Connolly, A

    Yip, C. W., Connolly, A. J., Vanden Berk, D. E., et al. 2004b, The Astronomical Journal, 128, 2603–2630, doi: 10.1086/425626

    work page doi:10.1086/425626

  55. [55]

    MUltiplexed Survey Telescope (MUST) Science White Paper I: Overview of Large-Scale Structure Cosmology in the Era of Stage-V Spectroscopic Surveys

    Zhao, C., Huang, S., He, M., et al. 2024, arXiv e-prints, arXiv:2411.07970, doi: 10.48550/arXiv.2411.07970

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2411.07970 2024

  56. [56]

    , keywords =

    Zhou, R., Dey, B., Newman, J. A., et al. 2023, The Astronomical Journal, 165, 58, doi: 10.3847/1538-3881/aca5fb

    work page doi:10.3847/1538-3881/aca5fb 2023

  57. [57]

    Zou, H., Zhou, X., Fan, X., et al. 2017, Publications of the Astronomical Society of the Pacific, 129, 064101, doi: 10.1088/1538-3873/aa65ba This paper was built using the Open Journal of As- trophysics LATEX template. The OJA is a journal which provides fast and easy peer review for new papers in the astro-phsectionofthearXiv, makingthereviewingpro- cess...

    work page doi:10.1088/1538-3873/aa65ba 2017