Deep Neural Network Classifier for Variable Stars with Novelty Detection Capability

Abstract

Common variable star classifiers are built only with the goal of producing the correct class labels, leaving much of the multi-task capability of deep neural networks unexplored. We present a periodic light curve classifier that combines a recurrent neural network autoencoder for unsupervised feature extraction and a dual-purpose estimation network for supervised classification and novelty detection. The estimation network optimizes a Gaussian mixture model in the reduced-dimension feature space, where each Gaussian component corresponds to a variable class. An estimation network with a basic structure of a single hidden layer attains a cross-validation classification accuracy of ~99%, on par with the conventional workhorses, random forest classifiers. With the addition of photometric features, the network is capable of detecting previously unseen types of variability with precision 0.90, recall 0.96, and an F1 score of 0.93. The simultaneous training of the autoencoder and estimation network is found to be mutually beneficial, resulting in faster autoencoder convergence, and superior classification and novelty detection performance. The estimation network also delivers adequate results even when optimized with pre-trained autoencoder features, suggesting that it can readily extend existing classifiers to provide added novelty detection capabilities.

9 pages, 3 figures, 2 tables. Accepted to ApJ Letters

References in corpus (19)

• LSST: from Science Drivers to Reference Design and Anticipated Data Products
• The Zwicky Transient Facility: System Overview, Performance, and First Results
• The Man Behind the Curtain: X-rays Drive the UV through NIR Variability in the 2013 AGN Outburst in NGC 2617
• A box-fitting algorithm in the search for periodic transits
• Rotation-invariant convolutional neural networks for galaxy morphology prediction
• The ASAS-SN Catalog of Variable Stars I: The Serendipitous Survey
• Identifying Exoplanets with Deep Learning: A Five Planet Resonant Chain around Kepler-80 and an Eighth Planet around Kepler-90
• On Machine-Learned Classification of Variable Stars with Sparse and Noisy Time-Series Data
• Probabilistic Random Forest: A machine learning algorithm for noisy datasets
• Random forest automated supervised classification of Hipparcos periodic variable stars
• Automating Discovery and Classification of Transients and Variable Stars in the Synoptic Survey Era
• Classifying Radio Galaxies with Convolutional Neural Network
• A recurrent neural network for classification of unevenly sampled variable stars
• Deep-HiTS: Rotation Invariant Convolutional Neural Network for Transient Detection
• The EPOCH Project: I. Periodic variable stars in the EROS-2 LMC database
• A Package for the Automated Classification of Periodic Variable Stars
• Deep multi-survey classification of variable stars
• Automated Classification of Periodic Variable Stars detected by the Wide-field Infrared Survey Explorer
• Effective Image Differencing with ConvNets for Real-time Transient Hunting

Cited by in corpus (16)

• The Automatic Learning for the Rapid Classification of Events (ALeRCE) Alert Broker
• Dimensionality Reduction of SDSS Spectra with Variational Autoencoders
• Astronomia ex machina: a history, primer, and outlook on neural networks in astronomy
• Double-, triple-line spectroscopic candidates in the LAMOST Medium-Resolution Spectroscopic Survey
• On Neural Architectures for Astronomical Time-series Classification with Application to Variable Stars
• A review of unsupervised learning in astronomy
• Imbalance Learning for Variable Star Classification
• ASTROMER: A transformer-based embedding for the representation of light curves
• Classifying Kepler light curves for 12,000 A and F stars using supervised feature-based machine learning
• Light curve classification with recurrent neural networks for GOTO: dealing with imbalanced data
• Applications of Machine Learning to Predicting Core-collapse Supernova Explosion Outcomes
• Deep Attention-Based Supernovae Classification of Multi-Band Light-Curves
• Classification of Periodic Variable Stars with Novel Cyclic-Permutation Invariant Neural Networks
• LSST Cadence Strategy Evaluations for AGN Time-series Data in Wide-Fast-Deep Field
• HyPhy: Deep Generative Conditional Posterior Mapping of Hydrodynamical Physics
• Heartbeat Stars Recognition Based on Recurrent Neural Networks: Method and Validation