A CNN adapted to time series for the classification of Supernovae

Abstract : Cosmologists are facing the problem of the analysis of a huge quantity of data when observing the sky. The methods used in cos-mology are, for the most of them, relying on astrophysical models, and thus, for the classification, they usually use a machine learning approach in two-steps, which consists in, first, extracting features , and second, using a classifier. In this paper, we are specifically studying the supernovae phenomenon and especially the binary classification "I.a supernovae versus not-I.a supernovae". We present two Convolutional Neural Networks (CNNs) defeating the current state-of-the-art. The first one is adapted to time series and thus to the treatment of supernovae light-curves. The second one is based on a Siamese CNN and is suited to the nature of data, i.e. their sparsity and their weak quantity (small learning database).
Complete list of metadatas

Cited literature [7 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-01968489
Contributor : Marc Chaumont <>
Submitted on : Wednesday, January 2, 2019 - 6:39:34 PM
Last modification on : Thursday, April 4, 2019 - 7:36:18 PM
Long-term archiving on : Wednesday, April 3, 2019 - 3:41:11 PM

File

IST_ELECTRONIC_IMAGING_Color_I...
Files produced by the author(s)

Identifiers

  • HAL Id : lirmm-01968489, version 1

Citation

Anthony Brunel, Johanna Pasquet, Jérôme Pasquet, Nancy Rodriguez, Frédéric Comby, et al.. A CNN adapted to time series for the classification of Supernovae. Electronic Imaging, Jan 2019, Burlingame, CA, United States. ⟨lirmm-01968489⟩

Share

Metrics

Record views

244

Files downloads

143