Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders

Pablo Huijse, Nicolas Astorga, Pablo Estévez, Giuliano Pignata

Resultado de la investigación: Conference contribution

Resumen

The Chilean Automatic Supernovae SEarch (CHASE) is a survey designed to detect early Supernovae. In this paper we explore deep autoencoders to obtain a compressed latent space for a large transient candidate database from the CHASE image difference pipeline. Compared to conventional methods, the latent variables obtained with variational au-toencoders preserve more information and are more discriminative towards real astronomical transients.

Idioma originalEnglish
Título de la publicación alojadaESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning
Editoriali6doc.com publication
Páginas321-326
Número de páginas6
ISBN (versión digital)9782875870476
EstadoPublished - 1 ene 2018
Evento26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2018 - Bruges, Belgium
Duración: 25 abr 201827 abr 2018

Serie de la publicación

NombreESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

Conference

Conference26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2018
PaísBelgium
CiudadBruges
Período25/04/1827/04/18

Huella dactilar

Pipelines
Supernovae

ASJC Scopus subject areas

  • Artificial Intelligence
  • Information Systems

Citar esto

Huijse, P., Astorga, N., Estévez, P., & Pignata, G. (2018). Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders. En ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (pp. 321-326). (ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning). i6doc.com publication.
Huijse, Pablo ; Astorga, Nicolas ; Estévez, Pablo ; Pignata, Giuliano. / Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders. ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com publication, 2018. pp. 321-326 (ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).
@inproceedings{57618e5024e34e4dbb357058d231f57a,
title = "Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders",
abstract = "The Chilean Automatic Supernovae SEarch (CHASE) is a survey designed to detect early Supernovae. In this paper we explore deep autoencoders to obtain a compressed latent space for a large transient candidate database from the CHASE image difference pipeline. Compared to conventional methods, the latent variables obtained with variational au-toencoders preserve more information and are more discriminative towards real astronomical transients.",
author = "Pablo Huijse and Nicolas Astorga and Pablo Est{\'e}vez and Giuliano Pignata",
year = "2018",
month = "1",
day = "1",
language = "English",
series = "ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",
publisher = "i6doc.com publication",
pages = "321--326",
booktitle = "ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning",

}

Huijse, P, Astorga, N, Estévez, P & Pignata, G 2018, Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders. En ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, i6doc.com publication, pp. 321-326, 26th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning, ESANN 2018, Bruges, Belgium, 25/04/18.

Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders. / Huijse, Pablo; Astorga, Nicolas; Estévez, Pablo; Pignata, Giuliano.

ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com publication, 2018. p. 321-326 (ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).

Resultado de la investigación: Conference contribution

TY - GEN

T1 - Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders

AU - Huijse, Pablo

AU - Astorga, Nicolas

AU - Estévez, Pablo

AU - Pignata, Giuliano

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The Chilean Automatic Supernovae SEarch (CHASE) is a survey designed to detect early Supernovae. In this paper we explore deep autoencoders to obtain a compressed latent space for a large transient candidate database from the CHASE image difference pipeline. Compared to conventional methods, the latent variables obtained with variational au-toencoders preserve more information and are more discriminative towards real astronomical transients.

AB - The Chilean Automatic Supernovae SEarch (CHASE) is a survey designed to detect early Supernovae. In this paper we explore deep autoencoders to obtain a compressed latent space for a large transient candidate database from the CHASE image difference pipeline. Compared to conventional methods, the latent variables obtained with variational au-toencoders preserve more information and are more discriminative towards real astronomical transients.

UR - http://www.scopus.com/inward/record.url?scp=85069480168&partnerID=8YFLogxK

M3 - Conference contribution

T3 - ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

SP - 321

EP - 326

BT - ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning

PB - i6doc.com publication

ER -

Huijse P, Astorga N, Estévez P, Pignata G. Latent representations of transient candidates from an astronomical image difference pipeline using Variational Autoencoders. En ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning. i6doc.com publication. 2018. p. 321-326. (ESANN 2018 - Proceedings, European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning).