Galaxy Spectra Networks (GaSNet) – III. Reconstructive pre-trained network for spectrum reconstruction, redshift estimate, and anomaly detection

ABSTRACT Classification of spectra (1) and anomaly detection (2) are fundamental steps to guarantee the highest accuracy in redshift measurements (3) in modern all-sky spectroscopic surveys. We introduce a new Galaxy Spectra Neural Network (GaSNet-III) model that utilizes neural networks to perform these three tasks simultaneously with high efficiency. Two different reconstruction networks – an autoencoder-like network and a U-Net – are used to reconstruct the rest-frame spectrum, which is then compared with the observed spectrum via a $\chi ^2$ metric across the entire type and redshift spaces to find the best-fitting solution. SDSS DR16 spectra are used as a reference data set to provide a fully self-consistent science test to show that our model achieves accuracy comparable to that of classical principal component analysis-based methods, and even better in some specific metrics, while maintaining significantly higher efficiency. In particular, the model achieves an average of $>98~{{\ \rm per\ cent}}$ classification accuracy across all classes, and redshift accuracies of over 99 per cent for stars, over 98 per cent for galaxies with errors of the order of $\mathcal {O}(10^{-4})$, and over 93 per cent for quasars with errors of the order of $\mathcal {O}(10^{-3})$. Tests on DESI spectra demonstrate that the model can generalize well to other surveys without retraining, with only a small degradation in performance. Furthermore, by comparing different peaks of $\chi ^2$ curves, we define a robustness measure that enables the identification of anomalous spectra. The GaSNet-III provides accurate and high-efficiency spectrum modelling to perform accurate redshift estimates and anomaly detection in vast data volumes from future spectroscopic sky surveys.

Authors

Institutions

Publication Details

Journal
Monthly Notices of the Royal Astronomical Society
Published
2025-09-15
DOI
https://doi.org/10.1093/mnras/staf1482
Citations
2
Primary Topic
Spectroscopy Techniques in Biomedical and Chemical Research
Type
article
Field-Weighted Citation Impact
3.27
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

Galaxy Spectra Networks (GaSNet) – III. Reconstructive pre-trained network for spectrum reconstruction, redshift estimate, and anomaly detection

N. R. Napolitano, Aurélien Verdier, F. E. Bauer, Roberto J. Assef et al.
2 citations
Monthly Notices of the Royal Astronomical Society
Spectroscopy Techniques in Biomedical and Chemical Research
3.27
article

Galaxy Spectra Networks (GaSNet) – III. Reconstructive pre-trained network for spectrum reconstruction, redshift estimate, and anomaly detection

N. R. Napolitano, Aurélien Verdier, F. E. Bauer, Roberto J. Assef, Jens-Kristian Krogager, Steve Ardern, Boudewijn F. Roukema, Fucheng Zhong, Caroline Heneka, J. Loveday, N. Bouché, Cristobál Sifón, Ricardo DeMarco, Alexander Fritz, L. P. Cassará
article en
2 citations

Abstract

ABSTRACT Classification of spectra (1) and anomaly detection (2) are fundamental steps to guarantee the highest accuracy in redshift measurements (3) in modern all-sky spectroscopic surveys. We introduce a new Galaxy Spectra Neural Network (GaSNet-III) model that utilizes neural networks to perform these three tasks simultaneously with high efficiency. Two different reconstruction networks – an autoencoder-like network and a U-Net – are used to reconstruct the rest-frame spectrum, which is then compared with the observed spectrum via a $\chi ^2$ metric across the entire type and redshift spaces to find the best-fitting solution. SDSS DR16 spectra are used as a reference data set to provide a fully self-consistent science test to show that our model achieves accuracy comparable to that of classical principal component analysis-based methods, and even better in some specific metrics, while maintaining significantly higher efficiency. In particular, the model achieves an average of $>98~{{\ \rm per\ cent}}$ classification accuracy across all classes, and redshift accuracies of over 99 per cent for stars, over 98 per cent for galaxies with errors of the order of $\mathcal {O}(10^{-4})$, and over 93 per cent for quasars with errors of the order of $\mathcal {O}(10^{-3})$. Tests on DESI spectra demonstrate that the model can generalize well to other surveys without retraining, with only a small degradation in performance. Furthermore, by comparing different peaks of $\chi ^2$ curves, we define a robustness measure that enables the identification of anomalous spectra. The GaSNet-III provides accurate and high-efficiency spectrum modelling to perform accurate redshift estimates and anomaly detection in vast data volumes from future spectroscopic sky surveys.

Monthly Notices of the Royal Astronomical SocietyVol. 543(1)
Université Claude Bernard Lyon 1 (FR), Centre National de la Recherche Scientifique (FR), Pontificia Universidad Católica de Valparaíso (CL), Universidad Andrés Bello (CL), Sun Yat-sen University (CN), University of Sussex (GB), University of Concepción (CL), University of Tarapacá (CL), Heidelberg University (DE), Astronomical Observatory of Capodimonte (IT), Centre de Recherche Astrophysique de Lyon (FR), Institute of Physics (AZ), Southern Marine Science and Engineering Guangdong Laboratory (Guangzhou) (CN), Pontificial Catholic University of Valparaiso (CL), Federico II University Hospital (IT), Nicolaus Copernicus Astronomical Center (PL), Otto Wagner Hospital (AT), École Polytechnique Fédérale de Lausanne (CH), University of Bath (GB), Heidelberg Institute for Theoretical Studies (DE), University of Naples Federico II (IT), National Institute for Astrophysics (IT), Universidad Diego Portales (CL)
Openalex Percentile: Top 7%
Spectroscopy Techniques in Biomedical and Chemical Research
3.27
DOIOpen Access PDFOpenAlex
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.