1. DELIGHT: Deep Learning Identification of Galaxy Hosts of Transients using Multi-resolution Images
- Author
-
Förster, Francisco, Arancibia, Alejandra M. Muñoz, Reyes, Ignacio, Gagliano, Alexander, Britt, Dylan, Cuellar-Carrillo, Sara, Figueroa-Tapia, Felipe, Polzin, Ava, Yousef, Yara, Arredondo, Javier, Rodríguez-Mancini, Diego, Correa-Orellana, Javier, Bayo, Amelia, Bauer, Franz E., Catelan, Márcio, Cabrera-Vives, Guillermo, Dastidar, Raya, Estévez, Pablo A., Pignata, Giuliano, Hernandez-Garcia, Lorena, Huijse, Pablo, Reyes, Esteban, Sánchez-Sáez, Paula, Ramirez, Mauricio, Grandón, Daniela, Pineda-García, Jonathan, Chabour-Barra, Francisca, and Silva-Farfán, Javier
- Subjects
Astrophysics - Instrumentation and Methods for Astrophysics - Abstract
We present DELIGHT, or Deep Learning Identification of Galaxy Hosts of Transients, a new algorithm designed to automatically and in real-time identify the host galaxies of extragalactic transients. The proposed algorithm receives as input compact, multi-resolution images centered at the position of a transient candidate and outputs two-dimensional offset vectors that connect the transient with the center of its predicted host. The multi-resolution input consists of a set of images with the same number of pixels, but with progressively larger pixel sizes and fields of view. A sample of \nSample galaxies visually identified by the ALeRCE broker team was used to train a convolutional neural network regression model. We show that this method is able to correctly identify both relatively large ($10\arcsec < r < 60\arcsec$) and small ($r \le 10\arcsec$) apparent size host galaxies using much less information (32 kB) than with a large, single-resolution image (920 kB). The proposed method has fewer catastrophic errors in recovering the position and is more complete and has less contamination ($< 0.86\%$) recovering the cross-matched redshift than other state-of-the-art methods. The more efficient representation provided by multi-resolution input images could allow for the identification of transient host galaxies in real-time, if adopted in alert streams from new generation of large etendue telescopes such as the Vera C. Rubin Observatory., Comment: Submitted to The Astronomical Journal on Aug 5th, 2022. Comments and suggestions are welcome
- Published
- 2022
- Full Text
- View/download PDF