Pith. sign in

REVIEW 6 cited by

Using Active Learning to Improve Quasar Identification for the DESI Spectra Processing Pipeline

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2505.01596 v2 pith:QXJRXDYU submitted 2025-05-02 astro-ph.IM astro-ph.CO

Using Active Learning to Improve Quasar Identification for the DESI Spectra Processing Pipeline

classification astro-ph.IM astro-ph.CO
keywords spectradatadesiquasarnetactivelearningpipelineweights
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

The Dark Energy Spectroscopic Instrument (DESI) survey uses an automatic spectral classification pipeline to classify spectra. QuasarNET is a convolutional neural network used as part of this pipeline originally trained using data from the Baryon Oscillation Spectroscopic Survey (BOSS). In this paper we implement an active learning algorithm to optimally select spectra to use for training a new version of the QuasarNET weights file using only DESI data, specifically to improve classification accuracy. This active learning algorithm includes a novel outlier rejection step using a Self-Organizing Map to ensure we label spectra representative of the larger quasar sample observed in DESI. We perform two iterations of the active learning pipeline, assembling a final dataset of 5600 labeled spectra, a small subset of the approx 1.3 million quasar targets in DESI's Data Release 1. When splitting the spectra into training and validation subsets we meet or exceed the previously trained weights file in completeness and purity calculated on the validation dataset with less than one tenth of the amount of training data. The new weights also more consistently classify objects in the same way when used on unlabeled data compared to the old weights file. In the process of improving QuasarNET's classification accuracy we discovered a systemic error in QuasarNET's redshift estimation and used our findings to improve our understanding of QuasarNET's redshifts.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

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

    astro-ph.CO 2025-03 accept novelty 7.0

    DESI DR2 BAO data exhibits 2.3 sigma tension with CMB in Lambda-CDM but prefers evolving dark energy (w0 > -1, wa < 0) at 3.1 sigma with CMB and 2.8-4.2 sigma when including supernovae.

  2. DESI DR2 Results IV: Alcock-Paczy\'nski Measurements from the Lyman Alpha Forest and Cosmological Constraints

    astro-ph.CO 2026-07 conditional novelty 6.0

    The full shape of DESI DR2 Lyman-alpha forest correlations constrains the distance ratio DM/DH at z=2.33 to 1.0%, twice as precise as BAO alone.

  3. DESI DR2 Results IV: Alcock-Paczy\'nski Measurements from the Lyman Alpha Forest and Cosmological Constraints

    astro-ph.CO 2026-07 accept novelty 6.0

    DESI DR2 Lyman-alpha forest full-shape correlations yield a 1% Alcock-Paczyński measurement at z=2.33 and 0.8% distance ratio constraints.

  4. Early results in the search for extreme coronal line emitters with the Dark Energy Spectroscopic Instrument

    astro-ph.HE 2026-01 conditional novelty 6.0

    A DESI early-data search found three tidal-disruption-event-linked extreme coronal line emitters, giving a galaxy-normalized rate of 5 (+5/-3) × 10^-6 galaxy^-1 yr^-1 at z ≈ 0.2.

  5. Validation of the DESI DR2 Ly$\alpha$ forest full-shape analysis

    astro-ph.CO 2026-07 conditional novelty 5.0

    The DESI DR2 Lyman-alpha full-shape analysis passes validation for BAO and Alcock-Paczynski parameters on 400 mocks and blinded data, while f-sigma-8 is rejected due to a roughly 10% mock bias.

  6. DESI DR2 Results I: Baryon Acoustic Oscillations from the Lyman Alpha Forest

    astro-ph.CO 2025-03 accept novelty 4.0

    DESI DR2 delivers 0.65% precision BAO measurements from the LyA forest at z_eff=2.33, with D_H/r_d = 8.632 ± 0.098 ± 0.026 and D_M/r_d = 38.99 ± 0.52 ± 0.12.