A Poisson Process AutoDecoder for X-ray Sources
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929698681913344 |
|---|---|
| author | Song, Yanke Villar, Victoria Ashley Martinez-Galarza, Juan Rafael Dillmann, Steven |
| author_facet | Song, Yanke Villar, Victoria Ashley Martinez-Galarza, Juan Rafael Dillmann, Steven |
| contents | X-ray observing facilities, such as the Chandra X-ray Observatory and the eROSITA, have detected millions of 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 Poisson Process AutoDecoder (PPAD). PPAD 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 | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_01627 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A Poisson Process AutoDecoder for X-ray Sources Song, Yanke Villar, Victoria Ashley Martinez-Galarza, Juan Rafael Dillmann, Steven Instrumentation and Methods for Astrophysics High Energy Astrophysical Phenomena Machine Learning Applications X-ray observing facilities, such as the Chandra X-ray Observatory and the eROSITA, have detected millions of 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 Poisson Process AutoDecoder (PPAD). PPAD 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. |
| title | A Poisson Process AutoDecoder for X-ray Sources |
| topic | Instrumentation and Methods for Astrophysics High Energy Astrophysical Phenomena Machine Learning Applications |
| url | https://arxiv.org/abs/2502.01627 |