Real-Time Detection of Anomalies in Large-Scale Transient Surveys

TL;DR

This paper introduces two real-time anomaly detection methods for transient light curves in large-scale surveys, using probabilistic neural networks and Bayesian models, to identify rare and unusual astronomical events efficiently.

Contribution

The paper presents two novel methods for real-time anomaly detection in transient surveys, demonstrating their effectiveness on Zwicky Transient Facility data and highlighting the advantages of the Bayesian approach.

Findings

Bayesian parametric model achieves high precision and recall for rare transients.

Neural network approach is less suitable for anomaly detection than the Bayesian model.

Anomaly detection performance improves as light curves evolve over time.

Abstract

New time-domain surveys, such as the Vera C. Rubin Observatory Legacy Survey of Space and Time (LSST), will observe millions of transient alerts each night, making standard approaches of visually identifying new and interesting transients infeasible. We present two novel methods of automatically detecting anomalous transient light curves in real-time. Both methods are based on the simple idea that if the light curves from a known population of transients can be accurately modelled, any deviations from model predictions are likely anomalies. The first modelling approach is a probabilistic neural network built using Temporal Convolutional Networks (TCNs) and the second is an interpretable Bayesian parametric model of a transient. We demonstrate our methods' ability to provide anomaly scores as a function of time on light curves from the Zwicky Transient Facility. We show that the flexibility of neural networks, the attribute that makes them such a powerful tool for many regression tasks, is what makes them less suitable for anomaly detection when compared with our parametric model. The parametric model is able to identify anomalies with respect to common supernova classes with high precision and recall scores, achieving area under the precision-recall curves (AUCPR) above 0.79 for most rare classes such as kilonovae, tidal disruption events, intermediate luminosity transients, and pair-instability supernovae. Our ability to identify anomalies improves over the lifetime of the light curves. Our framework, used in conjunction with transient classifiers, will enable fast and prioritised followup of unusual transients from new large-scale surveys.