Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation

Abstract

Since the variety of their light curve morphologies, the vast majority of the known heartbeat stars (HBSs) have been discovered by manual inspection. Machine learning, which has already been successfully applied to the classification of variable stars based on light curves, offers another possibility for the automatic detection of HBSs. We propose a novel feature extraction approach for HBSs. First, the orbital frequencies are calculated automatically according to the Fourier spectra of the light curves. Then, the amplitudes of the first 100 harmonics are extracted. Finally, these harmonics are normalized as feature vectors of the light curve. A training data set of synthetic light curves is constructed using ELLC, and their features are fed into recurrent neural networks (RNNs) for supervised learning, with the expected output being the eccentricity of these light curves. The performance of the RNNs is evaluated using a test data set of synthetic light curves, achieving 95$\%$ accuracy. When applied to known HBSs from the OGLE, Kepler, and TESS surveys, the networks achieve an average accuracy of 86$\%$. This method successfully identifies four new HBSs within the eclipsing binary catalog of Kirk et al. The use of orbital harmonics as features for HBSs proves to be a practical approach that significantly reduces the computational cost of neural networks. RNNs show excellent performance in recognizing this type of time series data. This method not only allows efficient identification of HBSs but can also be extended to recognize other types of periodic variable stars.

10pages,5figures

References in corpus (53)

• Kepler Eclipsing Binary Stars. VII. The Catalog of Eclipsing Binaries Found in the Entire Kepler Data-Set
• A Class of Eccentric Binaries with Dynamic Tidal Distortions Discovered with Kepler
• Testing the asteroseismic scaling relations for Red Giants with Eclipsing Binaries Observed by Kepler
• KIC 4544587: an Eccentric, Short Period Binary System with delta Sct Pulsations and Tidally Excited Modes
• ELLC - a fast, flexible light curve model for detached eclipsing binary stars and transiting exoplanets
• Heartbeat Stars, Tidally Excited Oscillations, and Resonance Locking
• Pulsating red giant stars in eccentric binary systems discovered from Kepler space-based photometry
• Alert Classification for the ALeRCE Broker System: The Light Curve Classifier
• A recurrent neural network for classification of unevenly sampled variable stars
• Tidal Excitation of Modes in Binary Systems with Applications to Binary Pulsars
• K2 Variable Catalogue II: Machine Learning Classification of Variable Stars and Eclipsing Binaries in K2 Fields 0-4
• Establishing the accuracy of asteroseismic mass and radius estimates of giant stars. I. Three eclipsing systems at [Fe/H]~ -0.3 and the need for a large high-precision sample
• KIC 3858884: a hybrid δ Sct pulsator in a highly eccentric eclipsing binary
• Deep Learning Classification in Asteroseismology Using an Improved Neural Network: Results on 15000 Kepler Red Giants and Applications to K2 and TESS Data
• Radial velocity monitoring of Kepler heartbeat stars
• Machine-learning Approaches to Exoplanet Transit Detection and Candidate Validation in Wide-field Ground-based Surveys
• Rapid Classification of TESS Planet Candidates with Convolutional Neural Networks
• Machine Learning Techniques for Stellar Light Curve Classification
• KIC 4142768: An Evolved Gamma Doradus/Delta Scuti Hybrid Pulsating Eclipsing Binary with Tidally Excited Oscillations
• KIC 8164262: a Heartbeat Star Showing Tidally Induced Pulsations with Resonant Locking
• Scalable End-to-end Recurrent Neural Network for Variable star classification
• KIC 3749404: A heartbeat Star with Rapid Apsidal Advance Indicative of a Tertiary Component
• KIC 9246715: The Double Red Giant Eclipsing Binary With Odd Oscillations
• Deep multi-survey classification of variable stars
• Massive heartbeat stars from TESS. I. TESS sectors 1-16
• On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
• Tidally Induced Pulsations in Kepler Eclipsing Binary KIC 3230227
• It takes a village to raise a tide: nonlinear multiple-mode coupling and mode identification in KOI-54
• Tidally Excited Oscillations in Heartbeat Binary Stars: Pulsation Phases and Mode Identification
• Detecting Solar-like Oscillations in Red Giants with Deep Learning
• HD 183648: a Kepler eclipsing binary with anomalous ellipsoidal variations and a pulsating component
• Accelerated Tidal Circularization Via Resonance Locking in KIC 8164262
• Classifying Kepler light curves for 12,000 A and F stars using supervised feature-based machine learning
• TESS Data for Asteroseismology (T'DA) Stellar Variability Classification Pipeline: Set-Up and Application to the Kepler Q9 Data
• Finding flares in Kepler and TESS data with recurrent deep neural networks
• Deep Neural Network Classifier for Variable Stars with Novelty Detection Capability
• HIDES spectroscopy of bright detached eclipsing binaries from the $Kepler$ field - I. Single-lined objects
• The Loudest Stellar Heartbeat: Characterizing the most extreme amplitude heartbeat star system
• A machine-learning photometric classifier for massive stars in nearby galaxies I. The method
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• Photometric Analysis of the OGLE Heartbeat Stars
• The OGLE Collection of Variable Stars. One Thousand Heartbeat Stars in the Galactic Bulge and Magellanic Clouds
• Light curve solutions of 12 eccentric Kepler binaries and analysis of their out-of-eclipse variability
• Identifying Light-curve Signals with a Deep Learning Based Object Detection Algorithm. II. A General Light Curve Classification Framework
• Theoretical investigation of the occurrence of tidally excited oscillations in massive eccentric binary systems
• Variable Star Classification Using Multi-View Metric Learning
• Detection of Tidally Excited Oscillations in Kepler Heartbeat Stars
• Light Curve Solutions of Ten Eccentric Kepler Binaries, Three of them with Tidally Induced Humps
• Application of Convolutional Neural Networks to time domain astrophysics. 2D image analysis of OGLE light curves
• Searching for orbital decay in a heartbeat star system KIC 3766353
• Exploring extreme brightness variations in blue supergiant MACHO 80.7443.1718: Evidence for companion-driven enhanced mass loss
• Twenty-three New Heartbeat Star Systems Discovered Based on TESS Data
• KIC 7914906: An Eclipsing Heartbeat Star with Tidally Excited Oscillations and Gamma Doradus/Delta Scuti Hybrid Pulsations

Cited by in corpus (1)

• The Variable Sky Through the OGLE Eye