A recurrent neural network for classification of unevenly sampled variable stars

TL;DR

This paper introduces a novel recurrent neural network that effectively classifies unevenly sampled variable stars by leveraging sampling times and noise properties, reducing the need for manual feature engineering.

Contribution

The authors develop an unsupervised autoencoding RNN that explicitly incorporates sampling times and noise, achieving competitive classification performance without extensive feature engineering.

Findings

Autoencoded features perform well across different surveys.

The model rivals best-in-class classification methods.

Networks can be used for forecasting and anomaly detection.

Abstract

Astronomical surveys of celestial sources produce streams of noisy time series measuring flux versus time ("light curves"). Unlike in many other physical domains, however, large (and source-specific) temporal gaps in data arise naturally due to intranight cadence choices as well as diurnal and seasonal constraints. With nightly observations of millions of variable stars and transients from upcoming surveys, efficient and accurate discovery and classification techniques on noisy, irregularly sampled data must be employed with minimal human-in-the-loop involvement. Machine learning for inference tasks on such data traditionally requires the laborious hand-coding of domain-specific numerical summaries of raw data ("features"). Here we present a novel unsupervised autoencoding recurrent neural network (RNN) that makes explicit use of sampling times and known heteroskedastic noise properties. When trained on optical variable star catalogs, this network produces supervised classification models that rival other best-in-class approaches. We find that autoencoded features learned on one time-domain survey perform nearly as well when applied to another survey. These networks can continue to learn from new unlabeled observations and may be used in other unsupervised tasks such as forecasting and anomaly detection.