A Neural Network-Based Search for Unmodeled Transients in LIGO-Virgo-KAGRA's Third Observing Run

TL;DR

This paper introduces a neural network-based search method that effectively detects unmodeled gravitational-wave transients in LIGO-Virgo-KAGRA data, including known CBCs and detector glitches, without relying on specific signal models.

Contribution

The paper presents the GWAK method, a novel neural network approach that captures physical features of signals to improve detection of unmodeled transients in gravitational-wave data.

Findings

Successfully detects known CBCs and glitches

Constructs a low-dimensional embedded space for signals

Enhances detection capabilities beyond existing pipelines

Abstract

This paper presents the results of a Neural Network (NN)-based search for short-duration gravitational-wave transients in data from the third observing run of LIGO, Virgo, and KAGRA. The search targets unmodeled transients with durations of milliseconds to a few seconds in the 30-1500 Hz frequency band, without assumptions about the incoming signal direction, polarization, or morphology. Using the Gravitational Wave Anomalous Knowledge (GWAK) method, three compact binary coalescences (CBCs) identified by existing pipelines are successfully detected, along with a range of detector glitches. The algorithm constructs a low-dimensional embedded space to capture the physical features of signals, enabling the detection of CBCs, detector glitches, and unmodeled transients. This study demonstrates GWAK's ability to enhance gravitational-wave searches beyond the limits of existing pipelines, laying the groundwork for future detection strategies.