A Poisson Process AutoDecoder for X-Ray Sources
X-ray observing facilities, such as the Chandra X-ray Observatory and the eROSITA, have detected over a million astronomical sources associated with high-energy phenomena. The arrival of photons as a function of time follows a Poisson process and can vary by orders-of-magnitude, presenting obstacles...
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IOP Publishing
2025-01-01
|
| Series: | The Astrophysical Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.3847/1538-4357/add72e |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1839624497527783424 |
|---|---|
| author | Yanke Song V. Ashley Villar Rafael Martínez-Galarza Steven Dillmann |
| author_facet | Yanke Song V. Ashley Villar Rafael Martínez-Galarza Steven Dillmann |
| author_sort | Yanke Song |
| collection | DOAJ |
| description | X-ray observing facilities, such as the Chandra X-ray Observatory and the eROSITA, have detected over a million astronomical sources associated with high-energy phenomena. The arrival of photons as a function of time follows a Poisson process and can vary by orders-of-magnitude, presenting obstacles for common tasks such as source classification, physical property derivation, and anomaly detection. Previous work has either failed to directly capture the Poisson nature of the data or only focuses on Poisson rate function reconstruction. In this work, we present the Poisson Process AutoDecoder (PPAD), which is a neural field decoder that maps fixed-length latent features to continuous Poisson rate functions across energy band and time via unsupervised learning. PPAD reconstructs the rate function and yields a representation at the same time. We demonstrate the efficacy of PPAD via reconstruction, regression, classification, and anomaly detection experiments using the Chandra Source Catalog. |
| format | Article |
| id | doaj-art-3d82ab3b8dda40ff9ba64b91a587711d |
| institution | Matheson Library |
| issn | 1538-4357 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IOP Publishing |
| record_format | Article |
| series | The Astrophysical Journal |
| spelling | doaj-art-3d82ab3b8dda40ff9ba64b91a587711d2025-07-18T13:55:35ZengIOP PublishingThe Astrophysical Journal1538-43572025-01-01988114310.3847/1538-4357/add72eA Poisson Process AutoDecoder for X-Ray SourcesYanke Song0https://orcid.org/0009-0008-6173-6931V. Ashley Villar1https://orcid.org/0000-0002-5814-4061Rafael Martínez-Galarza2https://orcid.org/0000-0002-5069-0324Steven Dillmann3https://orcid.org/0000-0002-4773-1463Department of Statistics, Harvard University , USACenter for Astrophysics—Harvard & Smithsonian , Cambridge, MA 02138, USA; The NSF AI Institute for Artificial Intelligence and Fundamental Interactions , USACenter for Astrophysics—Harvard & Smithsonian , Cambridge, MA 02138, USAInstitute of Computational and Mathematical Engineering, Stanford University, USAX-ray observing facilities, such as the Chandra X-ray Observatory and the eROSITA, have detected over a million astronomical sources associated with high-energy phenomena. The arrival of photons as a function of time follows a Poisson process and can vary by orders-of-magnitude, presenting obstacles for common tasks such as source classification, physical property derivation, and anomaly detection. Previous work has either failed to directly capture the Poisson nature of the data or only focuses on Poisson rate function reconstruction. In this work, we present the Poisson Process AutoDecoder (PPAD), which is a neural field decoder that maps fixed-length latent features to continuous Poisson rate functions across energy band and time via unsupervised learning. PPAD reconstructs the rate function and yields a representation at the same time. We demonstrate the efficacy of PPAD via reconstruction, regression, classification, and anomaly detection experiments using the Chandra Source Catalog.https://doi.org/10.3847/1538-4357/add72eMultivariate analysisAstrostatistics techniquesTime series analysisX-ray astronomy |
| spellingShingle | Yanke Song V. Ashley Villar Rafael Martínez-Galarza Steven Dillmann A Poisson Process AutoDecoder for X-Ray Sources The Astrophysical Journal Multivariate analysis Astrostatistics techniques Time series analysis X-ray astronomy |
| title | A Poisson Process AutoDecoder for X-Ray Sources |
| title_full | A Poisson Process AutoDecoder for X-Ray Sources |
| title_fullStr | A Poisson Process AutoDecoder for X-Ray Sources |
| title_full_unstemmed | A Poisson Process AutoDecoder for X-Ray Sources |
| title_short | A Poisson Process AutoDecoder for X-Ray Sources |
| title_sort | poisson process autodecoder for x ray sources |
| topic | Multivariate analysis Astrostatistics techniques Time series analysis X-ray astronomy |
| url | https://doi.org/10.3847/1538-4357/add72e |
| work_keys_str_mv | AT yankesong apoissonprocessautodecoderforxraysources AT vashleyvillar apoissonprocessautodecoderforxraysources AT rafaelmartinezgalarza apoissonprocessautodecoderforxraysources AT stevendillmann apoissonprocessautodecoderforxraysources AT yankesong poissonprocessautodecoderforxraysources AT vashleyvillar poissonprocessautodecoderforxraysources AT rafaelmartinezgalarza poissonprocessautodecoderforxraysources AT stevendillmann poissonprocessautodecoderforxraysources |