Variable star classification using multiview metric learning

Johnston, K.; Caballero-Nieves, S. M.; Petit, Véronique; Peter, Adrian M.; Haber, Rana

doi:10.1093/mnras/stz3165

Cited by 8 publications

(3 citation statements)

References 78 publications

(96 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Distinct approaches in RL to automate features extraction from astronomical time-series have already been introduced in a broad range of studies. Used techniques include unsupervised learning algorithms (Armstrong et al 2016), dimensionality reduction techniques, data transformations (Johnston et al 2020), autoencoders (Naul et al 2018) or dictionary learning (Pieringer et al 2019).…”

Section: Introductionmentioning

confidence: 99%

On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars

Jamal,

Bloom

2020

Preprint

View full text Add to dashboard Cite

Despite the utility of neural networks (NNs) for astronomical time-series classification, the proliferation of learning architectures applied to diverse datasets has thus far hampered a direct intercomparison of different approaches. Here we perform the first comprehensive study of variants of NN-based learning and inference for astronomical time-series, aiming to provide the community with an overview on relative performance and, hopefully, a set of best-in-class choices for practical implementations. In both supervised and self-supervised contexts, we study the effects of different time-series-compatible layer choices, namely the dilated temporal convolutional neural network (dTCNs), Long-Short Term Memory (LSTM) NNs, Gated Recurrent Units (GRUs) and temporal convolutional NNs (tCNNs). We also study the efficacy and performance of encoder-decoder (i.e., autoencoder) networks compared to direct classification networks, different pathways to include auxiliary (non-time-series) metadata, and different approaches to incorporate multi-passband data (i.e., multiple timeseries per source). Performance-applied to a sample of 17,604 variable stars from the MACHO survey across 10 imbalanced classes-is measured in training convergence time, classification accuracy, reconstruction error, and generated latent variables. We find that networks with Recurrent NN (RNNs) generally outperform dTCNs and, in many scenarios, yield to similar accuracy as tCNNs. In learning time and memory requirements, convolution-based layers are more performant. We conclude by discussing the advantages and limitations of deep architectures for variable star classification, with a particular eye towards next-generation surveys such as LSST, WFIRST and ZTF2.

show abstract

Section: Introductionmentioning

confidence: 99%

On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars

Jamal,

Bloom

2020

Preprint

View full text Add to dashboard Cite

show abstract

“…In particular, large-scale photometric surveys are producing light curves in numbers too large for humans to manually inspect and analyse. Considerable efforts have gone into using machine learning to classify light curves from large ground-based surveys (e.g., Carrasco-Davis et al 2019;Tsang & Schultz 2019;Johnston et al 2019a;Cabral et al 2020;Hosenie et al 2020;Jamal & Bloom 2020;Szklenár et al 2020;Bassi et al 2021;Zhang & Bloom 2021). Such techniques have also been applied to light curves from NASA's Kepler and K2 missions (e.g., Blomme et al 2010Blomme et al , 2011Debosscher et al 2011;Bass & Borne 2016;Armstrong et al 2016;Hon et al 2017Hon et al , 2018bJohnston et al 2019b;Kgoadi et al 2019;Le Saux et al 2019;Giles & Walkowicz 2020;Kuszlewicz et al 2020;Audenaert et al 2021;Paul & Chattopadhyay 2022).…”

Section: Introductionmentioning

confidence: 99%

Classifying Kepler light curves for 12,000 A and F stars using supervised feature-based machine learning

Barbara,

Bedding,

Fulcher

et al. 2022

Preprint

View full text Add to dashboard Cite

With the availability of large-scale surveys like Kepler and TESS, there is a pressing need for automated methods to classify light curves according to known classes of variable stars. We introduce a new algorithm for classifying light curves that compares 7000 time-series features to find those which most effectively classify a given set of light curves. We apply our method to Kepler light curves for stars with effective temperatures in the range 6500-10,000 K. We show that the sample can be meaningfully represented in an interpretable five-dimensional feature space that separates seven major classes of light curves (𝛿 Scuti stars, 𝛾 Doradus stars, RR Lyrae stars, rotational variables, contact eclipsing binaries, detached eclipsing binaries, and non-variables). We achieve a balanced classification accuracy of 82% on an independent test set of Kepler stars using a Gaussian mixture model classifier. We use our method to classify 12,000 Kepler light curves from Quarter 9 and provide a catalogue of the results. We further outline a confidence heuristic based on probability density with which to search our catalogue, and extract candidate lists of correctly-classified variable stars.

show abstract

“…Nun et al 2015), Fourierdecomposition (Kim & Bailer-Jones 2016) or color information (Miller et al 2015). The classifiers can be trained on manually designed (Pashchenko et al 2018;Hosenie et al 2019) or computer-selected features (Becker et al 2020;Johnston et al 2020) using known type of variable stars. Another opportunity to classify light curves is to use non-labeled data, which is called unsupervised learning.…”

Section: Introductionmentioning

confidence: 99%

Image-based Classification of Variable Stars: First Results from Optical Gravitational Lensing Experiment Data

Szklenár

Bódi

Tarczay-Nehéz

et al. 2020

ApJL

View full text Add to dashboard Cite

Recently, machine learning methods presented a viable solution for automated classification of image-based data in various research fields and business applications. Scientists require a fast and reliable solution to be able to handle the always growing enormous amount of data in astronomy. However, so far astronomers have been mainly classifying variable star light curves based on various pre-computed statistics and light curve parameters. In this work we use an image-based Convolutional Neural Network to classify the different types of variable stars. We used images of phase-folded light curves from the OGLE-III survey for training, validating and testing and used OGLE-IV survey as an independent data set for testing. After the training phase, our neural network was able to classify the different types between 80 and 99%, and 77-98% accuracy for OGLE-III and OGLE-IV, respectively.

show abstract

Variable star classification using multiview metric learning

Cited by 8 publications

References 78 publications

On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars

On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars

Classifying Kepler light curves for 12,000 A and F stars using supervised feature-based machine learning

Image-based Classification of Variable Stars: First Results from Optical Gravitational Lensing Experiment Data

Contact Info

Product

Resources

About