Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters
Inverse Compton (IC) emission associated with the nonthermal component of the intracluster medium (ICM) has been a long-sought phenomenon in cluster physics. Traditional spectral fitting often suffers from the degeneracy between the two-temperature thermal (2T) spectrum and the one-temperature plus...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2024-01-01
|
| Series: | The Astrophysical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4357/ad8888 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850127139963142144 |
|---|---|
| author | Sheng-Chieh Lin Yuanyuan Su Fabio Gastaldello Nathan Jacobs |
| author_facet | Sheng-Chieh Lin Yuanyuan Su Fabio Gastaldello Nathan Jacobs |
| author_sort | Sheng-Chieh Lin |
| collection | DOAJ |
| description | Inverse Compton (IC) emission associated with the nonthermal component of the intracluster medium (ICM) has been a long-sought phenomenon in cluster physics. Traditional spectral fitting often suffers from the degeneracy between the two-temperature thermal (2T) spectrum and the one-temperature plus IC power-law (1T+IC) spectrum. We present a semisupervised deep-learning approach to search for IC emission in galaxy clusters. We employ a conditional autoencoder (CAE), which is based on an autoencoder with latent representations trained to constrain the thermal parameters of the ICM. The algorithm is trained and tested using synthetic NuSTAR X-ray spectra with instrumental and astrophysical backgrounds included. The training data set only contains 2T spectra, which is more common than 1T+IC spectra. Anomaly detection is performed on the validation and test data sets consisting of 2T spectra as the normal set and 1T+IC spectra as anomalies. With a threshold anomaly score, chosen based on cross validation, our algorithm is able to identify spectra that contain an IC component in the test data set, with a balanced accuracy (BAcc) of 0.64, which outperforms traditional spectral fitting (BAcc = 0.55) and ordinary autoencoders (BAcc = 0.55). Traditional spectral fitting is better at identifying IC cases among true IC spectra (a better recall), while IC predictions made by CAE have a higher chance of being true IC cases (a better precision), demonstrating that they mutually complement each other. |
| format | Article |
| id | doaj-art-ac204b2bc3234df88ac43909211d041d |
| institution | OA Journals |
| issn | 1538-4357 |
| language | English |
| publishDate | 2024-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal |
| spelling | doaj-art-ac204b2bc3234df88ac43909211d041d2025-08-20T02:33:44ZengIOP PublishingThe Astrophysical Journal1538-43572024-01-01977217610.3847/1538-4357/ad8888Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy ClustersSheng-Chieh Lin0https://orcid.org/0000-0001-8178-8343Yuanyuan Su1https://orcid.org/0000-0002-3886-1258Fabio Gastaldello2https://orcid.org/0000-0002-9112-0184Nathan Jacobs3https://orcid.org/0000-0002-4242-8967Department of Physics and Astronomy, University of Kentucky , Lexington, KY, USADepartment of Physics and Astronomy, University of Kentucky , Lexington, KY, USAINAF , Istituto di Astrofisica Spaziale e Fisica Cosmica di Milano, Via A. Corti 12, 20133 Milano, ItalyDepartment of Computer Science & Engineering, Washington University in St. Louis , St. Louis, MO, USAInverse Compton (IC) emission associated with the nonthermal component of the intracluster medium (ICM) has been a long-sought phenomenon in cluster physics. Traditional spectral fitting often suffers from the degeneracy between the two-temperature thermal (2T) spectrum and the one-temperature plus IC power-law (1T+IC) spectrum. We present a semisupervised deep-learning approach to search for IC emission in galaxy clusters. We employ a conditional autoencoder (CAE), which is based on an autoencoder with latent representations trained to constrain the thermal parameters of the ICM. The algorithm is trained and tested using synthetic NuSTAR X-ray spectra with instrumental and astrophysical backgrounds included. The training data set only contains 2T spectra, which is more common than 1T+IC spectra. Anomaly detection is performed on the validation and test data sets consisting of 2T spectra as the normal set and 1T+IC spectra as anomalies. With a threshold anomaly score, chosen based on cross validation, our algorithm is able to identify spectra that contain an IC component in the test data set, with a balanced accuracy (BAcc) of 0.64, which outperforms traditional spectral fitting (BAcc = 0.55) and ordinary autoencoders (BAcc = 0.55). Traditional spectral fitting is better at identifying IC cases among true IC spectra (a better recall), while IC predictions made by CAE have a higher chance of being true IC cases (a better precision), demonstrating that they mutually complement each other.https://doi.org/10.3847/1538-4357/ad8888Galaxy clustersIntracluster mediumX-ray astronomyAstronomy data analysis |
| spellingShingle | Sheng-Chieh Lin Yuanyuan Su Fabio Gastaldello Nathan Jacobs Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters The Astrophysical Journal Galaxy clusters Intracluster medium X-ray astronomy Astronomy data analysis |
| title | Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters |
| title_full | Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters |
| title_fullStr | Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters |
| title_full_unstemmed | Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters |
| title_short | Semisupervised Learning for Detecting Inverse Compton Emission in Galaxy Clusters |
| title_sort | semisupervised learning for detecting inverse compton emission in galaxy clusters |
| topic | Galaxy clusters Intracluster medium X-ray astronomy Astronomy data analysis |
| url | https://doi.org/10.3847/1538-4357/ad8888 |
| work_keys_str_mv | AT shengchiehlin semisupervisedlearningfordetectinginversecomptonemissioningalaxyclusters AT yuanyuansu semisupervisedlearningfordetectinginversecomptonemissioningalaxyclusters AT fabiogastaldello semisupervisedlearningfordetectinginversecomptonemissioningalaxyclusters AT nathanjacobs semisupervisedlearningfordetectinginversecomptonemissioningalaxyclusters |