Sparse logistic regression for RR Lyrae vs binaries classification

Abstract

RR Lyrae (RRL) are old, low-mass radially pulsating variable stars in their core helium burning phase. They are popular stellar tracers and primary distance indicators, since they obey to well defined period-luminosity relations in the near-infrared regime. Their photometric identification is not trivial, indeed, RRL samples can be contaminated by eclipsing binaries, especially in large datasets produced by fully automatic pipelines. Interpretable machine-learning approaches for separating eclipsing binaries from RRL are thus needed. Ideally, they should be able to achieve high precision in identifying RRL while generalizing to new data from different instruments. In this paper, we train a simple logistic regression classifier on Catalina Sky Survey (CSS) light curves. It achieves a precision of 87% at 78% recall for the RRL class on unseen CSS light curves. It generalizes on out-of-sample data (ASAS/ASAS-SN light curves) with a precision of 85% at 96% recall. We also considered a L1-regularized version of our classifier, which reaches 90% sparsity in the light-curve features with a limited trade-off in accuracy on our CSS validation set and -- remarkably -- also on the ASAS/ASAS-SN light curve test set. Logistic regression is natively interpretable, and regularization allows us to point out the parts of the light curves that matter the most in classification. We thus achieved both good generalization and full interpretability.

arXiv commentTo appear on The Astrophysical Journal. 13 pages, 8 figures, 3 tables

References in corpus (35)

• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• Gaia Data Release 1. Summary of the astrometric, photometric, and survey properties
• The Man Behind the Curtain: X-rays Drive the UV through NIR Variability in the 2013 AGN Outburst in NGC 2617
• The Zwicky Transient Facility: Data Processing, Products, and Archive
• The Catalina Surveys Periodic Variable Star Catalog
• SciANN: A Keras/Tensorflow wrapper for scientific computations and physics-informed deep learning using artificial neural networks
• The RR Lyrae Period-Luminosity Relation. I. Theoretical Calibration
• Probing the Outer Galactic halo with RR Lyrae from the Catalina Surveys
• Automated supervised classification of variable stars I. Methodology
• CMU DeepLens: Deep Learning For Automatic Image-based Galaxy-Galaxy Strong Lens Finding
• The ASAS-SN Catalog of Variable Stars III: Variables in the Southern TESS Continuous Viewing Zone
• A pulsational approach to near infrared and visual magnitudes of RR Lyrae stars
• The Carnegie-Chicago Hubble Program. I. An Independent Approach to the Extragalactic Distance Scale Using only Population II Distance Indicators
• Discovery of $\sim$ 9,000 new RR Lyrae in the Southern Catalina Surveys
• Evidence for a Milky Way Tidal Stream Reaching Beyond 100 kpc
• Deep learning from 21-cm tomography of the Cosmic Dawn and Reionization
• On the distance of the globular cluster M4 (NGC 6121) using RR Lyrae stars: I. optical and near-infrared Period-Luminosity and Period-Wesenheit relations
• Analysis of RR Lyrae Stars in the Northern Sky Variability Survey
• Old-Aged Stellar Population Distance Indicators
• On the Use of Field RR Lyrae as Galactic Probes. II. A new $\Delta$S calibration to estimate their metallicity
• Machine Learning Techniques for Stellar Light Curve Classification
• Scalable End-to-end Recurrent Neural Network for Variable star classification
• Deep multi-survey classification of variable stars
• The VVV Survey RR Lyrae Population in the Galactic Centre Region
• On the Use of Field RR Lyrae as Galactic Probes. III. The $\alpha$-element abundances
• On the use of field RR Lyrae as Galactic probes: IV. New insights into and around the Oosterhoff dichotomy
• Deriving star cluster parameters with convolutional neural networks. I. Age, mass, and size
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• Variable star classification with a Multiple-Input Neural Network
• Uncertain classification of Variable Stars: handling observational GAPS and noise
• Interpreting automatic AGN classifiers with saliency maps
• Predicting the Stability of Hierarchical Triple Systems with Convolutional Neural Networks
• Semi-Supervised Classification and Clustering Analysis for Variable Stars
• Measuring the spectral index of turbulent gas with deep learning from projected density maps
• Predicting Images for the Dynamics Of stellar Clusters (π-DOC): a deep learning framework to predict mass, distance and age of globular clusters