Localization of gravitational waves using machine learning

TL;DR

This paper introduces a machine learning approach for rapid and potentially more accurate localization of gravitational wave sources, enhancing multi-messenger astronomy by integrating data from multiple detectors.

Contribution

It proposes a novel machine learning method combining existing techniques for fast gravitational wave sky localization, including the use of a temporal convolutional network.

Findings

Machine learning provides faster localization than matched filtering.

Adding KAGRA improves localization accuracy.

The method successfully localizes sources using four detectors.

Abstract

An observation of gravitational waves is a trigger of the multi-messenger search of an astronomical event. A combination of the data from two or three gravitational wave telescopes indicates the location of a source and low-latency data analysis is key to transferring the information to other telescopes sensitive at different wavelengths. In contrast to the current method, which relies on the matched-filtering technique, we proposed the use of machine learning that is much faster and possibly more accurate than matched filtering. Our machine-learning method is a combination of the method proposed by Chatterjee {\it et al.} and a method using the temporal convolutional network. We demonstrate the sky localization of a gravitational-wave source using four telescopes: LIGO H1, LIGO L1, Virgo, and KAGRA, and compare the result in the case without KAGRA to examine the positive influence of having the fourth telescope in the global gravitational-wave network.