A Deep Learning Approach for Active Anomaly Detection of Extragalactic Transients

TL;DR

This paper introduces an unsupervised deep learning method using a variational recurrent autoencoder and anomaly scoring to identify rare extragalactic transients in real-time from large astronomical datasets, aiding follow-up observations.

Contribution

It presents a novel unsupervised deep learning pipeline that effectively detects rare astrophysical transients from unlabeled, multivariate, and aperiodic data in real time.

Findings

Successfully ranks rarer transients as more anomalous.

Identifies a high-purity sample (~95%) of rare transients.

Detects transients as anomalous before peak brightness.

Abstract

There is a shortage of multi-wavelength and spectroscopic followup capabilities given the number of transient and variable astrophysical events discovered through wide-field, optical surveys such as the upcoming Vera C. Rubin Observatory. From the haystack of potential science targets, astronomers must allocate scarce resources to study a selection of needles in real time. Here we present a variational recurrent autoencoder neural network to encode simulated Rubin Observatory extragalactic transient events using 1% of the PLAsTiCC dataset to train the autoencoder. Our unsupervised method uniquely works with unlabeled, real time, multivariate and aperiodic data. We rank 1,129,184 events based on an anomaly score estimated using an isolation forest. We find that our pipeline successfully ranks rarer classes of transients as more anomalous. Using simple cuts in anomaly score and uncertainty, we identify a pure (~95% pure) sample of rare transients (i.e., transients other than Type Ia, Type II and Type Ibc supernovae) including superluminous and pair-instability supernovae. Finally, our algorithm is able to identify these transients as anomalous well before peak, enabling real-time follow up studies in the era of the Rubin Observatory.