GWSkyNet: a real-time classifier for public gravitational-wave candidates

TL;DR

GWSkyNet is a real-time neural network framework that rapidly classifies gravitational-wave candidates as astrophysical or artefacts using publicly available alert data, improving the speed and accuracy of follow-up observations.

Contribution

The paper introduces GWSkyNet, a novel convolutional neural network that classifies gravitational-wave candidates in real-time using only public alert data, enhancing detection reliability.

Findings

Achieves 93.5% accuracy on test data

Operates in real-time for prompt follow-up

Uses sky maps and metadata for classification

Abstract

The rapid release of accurate sky localization for gravitational-wave candidates is crucial for multi-messenger observations. During the third observing run of Advanced LIGO and Advanced Virgo, automated gravitational-wave alerts were publicly released within minutes of detection. Subsequent inspection and analysis resulted in the eventual retraction of a fraction of the candidates. Updates could be delayed by up to several days, sometimes issued during or after exhaustive multi-messenger followup campaigns. We introduce GWSkyNet, a real-time framework to distinguish between astrophysical events and instrumental artefacts using only publicly available information from the LIGO-Virgo open public alerts. This framework consists of a non-sequential convolutional neural network involving sky maps and metadata. GWSkyNet achieves a prediction accuracy of 93.5% on a testing data set.