Application of Convolutional Neural Networks to time domain astrophysics. 2D image analysis of OGLE light curves
Open Access
- 1 November 2024
- journal article
- research article
- Published by EDP Sciences in Astronomy & Astrophysics
- Vol. 691, A106
Abstract
In recent years the amount of publicly available astronomical data has increased exponentially, with a remarkable example being large-scale multiepoch photometric surveys. This wealth of data poses challenges to the classical methodologies commonly employed in the study of variable objects. As a response, deep learning techniques are increasingly being explored to effectively classify, analyze, and interpret these large datasets. In this paper we use two-dimensional histograms to represent Optical Gravitational Lensing Experiment (OGLE) phasefolded light curves as images. We use a Convolutional Neural Network (CNN) to classify variable objects within eight different categories (from now on labels): Classical Cepheid (CEP), RR Lyrae (RR), Long Period Variable (LPV), Miras (M), Ellipsoidal Binary (ELL), Delta Scuti (DST), Eclipsing Binary (E), and spurious class with Incorrect Periods (Rndm). We set up different training sets to train the same CNN architecture in order to characterize the impact of the training. The training sets were built from the same source of labels but different filters and balancing techniques were applied. Namely: Undersampling (U), Data Augmentation (DA), and Batch Balancing (BB). The best performance was achieved with the BB approach and a training sample size of sim 370000 stars. Regarding computational performance, the image representation production rate is of sim 76 images per core per second, and the time to predict is sim 60$\ s $ per star. The accuracy of the classification improves from sim 92<!PCT!>, when based only on the CNN, to sim 98<!PCT!> when the results of the CNN are combined with the period and amplitude features in a two step approach. This methodology achieves comparable results with previous studies but with two main advantages: the identification of miscalculated periods and the improvement in computational time cost.Keywords
All Related Versions
Funding Information
- Agencia Nacional de Investigación y Desarrollo (21211323)
- Corporación de Fomento de la Producción (18ENI2-104235)
- Fondo Nacional de Desarrollo Científico, Tecnológico y de Innovación Tecnológica (1211370)
- Deutsche Forschungsgemeinschaft (EXC 2094 - 390783311)
This publication has 54 references indexed in Scilit:
- A review of convolutional neural network architectures and their optimizationsArtificial Intelligence Review, 2022
- A survey on deep learning for big dataInformation Fusion, 2018
- Temporal Convolutional Networks: A Unified Approach to Action Segmentation2016
- Pulsating Stars2015
- VISTA Variables in the Via Lactea (VVV): The public ESO near-IR variability survey of the Milky WayNew Astronomy, 2010
- Accurate masses and radii of normal stars: modern results and applicationsThe Astronomy and Astrophysics Review, 2009
- Understanding Variable Stars2007
- COROT: A minisat for pionnier science, asteroseismology and planets findingAdvances In Space Research, 2003
- Random ForestsMachine Learning, 2001
- Least-squares frequency analysis of unequally spaced dataAstrophysics and Space Science, 1976