StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys
Extracting precise stellar labels is crucial for large spectroscopic surveys like the Sloan Digital Sky Survey (SDSS) and APOGEE. In this paper, we report the newest implementation of StellarGAN, a data-driven method based on generative adversarial networks (GANs). Using 1D operators like convolutio...
Main Authors: | , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
IOP Publishing
2024-01-01
|
Series: | The Astrophysical Journal Supplement Series |
Subjects: | |
Online Access: | https://doi.org/10.3847/1538-4365/ad29ef |
_version_ | 1797234802603065344 |
---|---|
author | Wei Liu Shuo Cao Xian-Chuan Yu Meng Zhu Marek Biesiada Jiawen Yao Minghao Du |
author_facet | Wei Liu Shuo Cao Xian-Chuan Yu Meng Zhu Marek Biesiada Jiawen Yao Minghao Du |
author_sort | Wei Liu |
collection | DOAJ |
description | Extracting precise stellar labels is crucial for large spectroscopic surveys like the Sloan Digital Sky Survey (SDSS) and APOGEE. In this paper, we report the newest implementation of StellarGAN, a data-driven method based on generative adversarial networks (GANs). Using 1D operators like convolution, the 2D GAN is modified into StellarGAN. This allows it to learn the relevant features of 1D stellar spectra without needing labels for specific stellar types. We test the performance of StellarGAN on different stellar spectra trained on SDSS and APOGEE data sets. Our result reveals that StellarGAN attains the highest overall F1-score on SDSS data sets (F1-score = 0.82, 0.77, 0.74, 0.53, 0.51, 0.61, and 0.55, for O-type, B-type, A-type, F-type, G-type, K-type, and M-type stars) when the signal-to-noise ratio (S/N) is low (90% of the spectra have an S/N < 50), with 1% of labeled spectra used for training. Using 50% of the labeled spectral data for training, StellarGAN consistently demonstrates performance that surpasses or is comparable to that of other data-driven models, as evidenced by the F1-scores of 0.92, 0.77, 0.77, 0.84, 0.84, 0.80, and 0.67. In the case of APOGEE (90% of the spectra have an S/N < 500), our method is also superior regarding its comprehensive performance (F1-score = 0.53, 0.60, 0.56, 0.56, and 0.78 for A-type, F-type, G-type, K-type, and M-type stars) with 1% of labeled spectra for training, manifesting its learning ability out of a limited number of labeled spectra. Our proposed method is also applicable to other types of data that need to be classified (such as gravitational-wave signals, light curves, etc.). |
first_indexed | 2024-04-24T16:37:51Z |
format | Article |
id | doaj.art-f14846ec6fbc41f5a8ff39e4db6f3c06 |
institution | Directory Open Access Journal |
issn | 0067-0049 |
language | English |
last_indexed | 2024-04-24T16:37:51Z |
publishDate | 2024-01-01 |
publisher | IOP Publishing |
record_format | Article |
series | The Astrophysical Journal Supplement Series |
spelling | doaj.art-f14846ec6fbc41f5a8ff39e4db6f3c062024-03-29T13:21:11ZengIOP PublishingThe Astrophysical Journal Supplement Series0067-00492024-01-0127125310.3847/1538-4365/ad29efStellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky SurveysWei Liu0Shuo Cao1https://orcid.org/0000-0002-8870-981XXian-Chuan Yu2Meng Zhu3Marek Biesiada4https://orcid.org/0000-0003-1308-7304Jiawen Yao5Minghao Du6Institute for Frontiers in Astronomy and Astrophysics, Beijing Normal University , Beijing 102206, People's Republic of China ; caoshuo@bnu.edu.cn; School of Artificial Intelligence, Beijing Normal University , Beijing 100875, People's Republic of China ; yuxianchuan@163.com; State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, People's Republic of ChinaInstitute for Frontiers in Astronomy and Astrophysics, Beijing Normal University , Beijing 102206, People's Republic of China ; caoshuo@bnu.edu.cn; Department of Astronomy, Beijing Normal University , Beijing 100875, People's Republic of ChinaSchool of Artificial Intelligence, Beijing Normal University , Beijing 100875, People's Republic of China ; yuxianchuan@163.comSchool of Artificial Intelligence, Beijing Normal University , Beijing 100875, People's Republic of China ; yuxianchuan@163.comNational Centre for Nuclear Research , Pasteura 7, 02-093, Warsaw, PolandDepartment of Astronomy, Beijing Normal University , Beijing 100875, People's Republic of ChinaDepartment of Astronomy, Beijing Normal University , Beijing 100875, People's Republic of ChinaExtracting precise stellar labels is crucial for large spectroscopic surveys like the Sloan Digital Sky Survey (SDSS) and APOGEE. In this paper, we report the newest implementation of StellarGAN, a data-driven method based on generative adversarial networks (GANs). Using 1D operators like convolution, the 2D GAN is modified into StellarGAN. This allows it to learn the relevant features of 1D stellar spectra without needing labels for specific stellar types. We test the performance of StellarGAN on different stellar spectra trained on SDSS and APOGEE data sets. Our result reveals that StellarGAN attains the highest overall F1-score on SDSS data sets (F1-score = 0.82, 0.77, 0.74, 0.53, 0.51, 0.61, and 0.55, for O-type, B-type, A-type, F-type, G-type, K-type, and M-type stars) when the signal-to-noise ratio (S/N) is low (90% of the spectra have an S/N < 50), with 1% of labeled spectra used for training. Using 50% of the labeled spectral data for training, StellarGAN consistently demonstrates performance that surpasses or is comparable to that of other data-driven models, as evidenced by the F1-scores of 0.92, 0.77, 0.77, 0.84, 0.84, 0.80, and 0.67. In the case of APOGEE (90% of the spectra have an S/N < 500), our method is also superior regarding its comprehensive performance (F1-score = 0.53, 0.60, 0.56, 0.56, and 0.78 for A-type, F-type, G-type, K-type, and M-type stars) with 1% of labeled spectra for training, manifesting its learning ability out of a limited number of labeled spectra. Our proposed method is also applicable to other types of data that need to be classified (such as gravitational-wave signals, light curves, etc.).https://doi.org/10.3847/1538-4365/ad29efStellar classificationStellar spectral types |
spellingShingle | Wei Liu Shuo Cao Xian-Chuan Yu Meng Zhu Marek Biesiada Jiawen Yao Minghao Du StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys The Astrophysical Journal Supplement Series Stellar classification Stellar spectral types |
title | StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys |
title_full | StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys |
title_fullStr | StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys |
title_full_unstemmed | StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys |
title_short | StellarGAN: Classifying Stellar Spectra with Generative Adversarial Networks in SDSS and APOGEE Sky Surveys |
title_sort | stellargan classifying stellar spectra with generative adversarial networks in sdss and apogee sky surveys |
topic | Stellar classification Stellar spectral types |
url | https://doi.org/10.3847/1538-4365/ad29ef |
work_keys_str_mv | AT weiliu stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys AT shuocao stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys AT xianchuanyu stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys AT mengzhu stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys AT marekbiesiada stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys AT jiawenyao stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys AT minghaodu stellarganclassifyingstellarspectrawithgenerativeadversarialnetworksinsdssandapogeeskysurveys |