Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection<sup>∗</sup>

Cabrera-Vives, G.; Reyes, I.; Förster, F.; Estévez, P. A.; Maureira, J. C.

doi:10.3847/1538-4357/836/1/97

Cited by 115 publications

(70 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Dieleman et al (2015) won the Galaxy Zoo Challenge by applying convolutional neural networks to galaxy morphology classification. Cabrera-Vives et al (2017) used the same ideas to classify transients in the HiTS survey, surpassing results over an RF. Aguirre et al (2018) developed a convolutional neural net-work to classify variable stars from multiple catalogues in a single model.…”

Section: Introductionmentioning

confidence: 99%

Scalable end-to-end recurrent neural network for variable star classification

Becker

Pichara

Catelan

et al. 2020

Monthly Notices of the Royal Astronomical Society

View full text Add to dashboard Cite

During the last decade, considerable effort has been made to perform automatic classification of variable stars using machine learning techniques. Traditionally, light curves are represented as a vector of descriptors or features used as input for many algorithms. Some features are computationally expensive, cannot be updated quickly and hence for large datasets such as the LSST cannot be applied. Previous work has been done to develop alternative unsupervised feature extraction algorithms for light curves, but the cost of doing so still remains high. In this work, we propose an end-to-end algorithm that automatically learns the representation of light curves that allows an accurate automatic classification. We study a series of deep learning architectures based on Recurrent Neural Networks and test them in automated classification scenarios. Our method uses minimal data preprocessing, can be updated with a low computational cost for new observations and light curves, and can scale up to massive datasets. We transform each light curve into an input matrix representation whose elements are the differences in time and magnitude, and the outputs are classification probabilities. We test our method in three surveys: OGLE-III, Gaia and WISE. We obtain accuracies of about 95% in the main classes and 75% in the majority of subclasses. We compare our results with the Random Forest classifier and obtain competitive accuracies while being faster and scalable. The analysis shows that the computational complexity of our approach grows up linearly with the light curve size, while the traditional approach cost grows as N log (N).

show abstract

Section: Introductionmentioning

confidence: 99%

Scalable end-to-end recurrent neural network for variable star classification

Becker

Pichara

Catelan

et al. 2020

Monthly Notices of the Royal Astronomical Society

View full text Add to dashboard Cite

show abstract

“…The data used for this work were processed and reduced in the same way as done for the 2014A campaign (see Förster et al 2016 andCabrera-Vives et al 2017), meaning that we had astrometry and photometry of moving objects and a probability for each of them of being real or bogus obtained using deep learning.…”

Section: Data Processingmentioning

confidence: 99%

Asteroids’ Size Distribution and Colors from HITS

Peña

Fuentes

Förster

et al. 2020

Self Cite

View full text Add to dashboard Cite

We report the observations of solar system objects during the 2015 campaign of the High cadence Transient Survey (HiTS). We found 5740 bodies (mostly Main Belt asteroids), 1203 of which were detected in different nights and in g and r . Objects were linked in the barycenter system and their orbital parameters were computed assuming Keplerian motion. We identified 6 near Earth objects, 1738 Main Belt asteroids and 4 Trans-Neptunian objects. We did not find a g −r color-size correlation for 14 < H g < 18 (1 < D < 10 km) asteroids. We show asteroids' colors are disturbed by HiTS' 1.6 hour cadence and estimate that observations should be separated by at most 14 minutes to avoid confusion in future wide-field surveys like LSST. The size distribution for the Main Belt objects can be characterized as a simple power law with slope ∼ 0.9 , steeper than in any other survey, while data from HiTS 2014's campaign is consistent with previous ones (slopes ∼ 0.68 at the bright end and ∼ 0.34 at the faint end). This difference is likely due to the ecliptic distribution of the Main Belt since 2015's campaign surveyed farther from the ecliptic than did 2014's and most previous surveys.

show abstract

“…Considerable efforts are made to classify (often irregularly spaced) light curves to identify transients and variable types in Big Data collections of light curves using training sets and machine learning techniques [e.g., [55][56][57][58][59][60][61]]. Autoregressive modeling results may be useful features to help discriminate variable classes.…”

Section: Classifying Time Seriesmentioning

confidence: 99%

Autoregressive Times Series Methods for Time Domain Astronomy

2018

View full text Add to dashboard Cite

Celestial objects exhibit a wide range of variability in brightness at different wavebands. Surprisingly, the most common methods for characterizing time series in statistics-parametric autoregressive modeling-are rarely used to interpret astronomical light curves. We review standard ARMA, ARIMA, and ARFIMA (autoregressive moving average fractionally integrated) models that treat short-memory autocorrelation, long-memory 1/f α "red noise," and nonstationary trends. Though designed for evenly spaced time series, moderately irregular cadences can be treated as evenly-spaced time series with missing data. Fitting algorithms are efficient and software implementations are widely available. We apply ARIMA models to light curves of four variable stars, discussing their effectiveness for different temporal characteristics. A variety of extensions to ARIMA are outlined, with emphasis on recently developed continuous-time models like CARMA and CARFIMA designed for irregularly spaced time series. Strengths and weakness of ARIMA-type modeling for astronomical data analysis and astrophysical insights are reviewed.Keywords: time domain astronomy, irregularly sampled time series, variable stars, quasars, statistical methods, time series analysis, autoregressive modeling, ARIMA THE VARIABILITY OF COSMIC POPULATIONSExcept for five roving planets and an occasional comet or nova, the nighttime sky seems immutable to the human eye. The pattern and brightness of stars appears unchanging as from our childhood to old age. Myths from ancient Egyptian, Greek, and Australian Aboriginal cultures suggest that a few stars (such as Algol, Mira, and Aldeberan) were recognized as variables [1,2]. As telescopic studies proliferated from the seventeenth through twenty first centuries, more variable stars were found with a wide range of characteristics. Some are periodic due to pulsations, rotationally modulated spots, or eclipses of binary companions. Others vary in irregular ways from magnetic flares, eruptions, pulsations, accretion of gas from companions, and most spectacularly, nova and supernova explosions. Ten thousand stars in two dozen categories were cataloged by Kukarkin and Parenago [3]; this catalog now has over 50,000 stars with >100 classes [4]. NASA's Kepler mission has recently shown that most ordinary stars are variable when observed with ∼0.001% accuracy and dense cadences [5].The study of celestial objects with variable brightness has broadened hugely in recent decades, emerging as a recognized discipline called "time domain astronomy" [6]. The brightest sources in the X-ray and gamma-ray sky are highly variable, typically from accretion of gas onto neutron stars and black holes. Timescales range from milliseconds to decades with a bewildering range of

show abstract

Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection^∗

Cited by 115 publications

References 24 publications

Scalable end-to-end recurrent neural network for variable star classification

Scalable end-to-end recurrent neural network for variable star classification

Asteroids’ Size Distribution and Colors from HITS

Autoregressive Times Series Methods for Time Domain Astronomy

Contact Info

Product

Resources

About

Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection∗

Cited by 115 publications

References 24 publications

Scalable end-to-end recurrent neural network for variable star classification

Scalable end-to-end recurrent neural network for variable star classification

Asteroids’ Size Distribution and Colors from HITS

Autoregressive Times Series Methods for Time Domain Astronomy

Contact Info

Product

Resources

About

Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection^∗