Abstract:Electromagnetic follow-up observations of gravitational wave events offer critical insights and provide significant scientific gain from this new class of astrophysical transients. Accurate identification of gravitational wave candidates and rapid release of sky localization information are crucial for the success of these electromagnetic follow-up observations. However, searches for gravitational wave candidates in real time suffer a non-negligible false alarm rate. By leveraging the sky localization information and other metadata associated with gravitational wave candidates, GWSkyNet, a machine learning classifier developed by Cabero et al. (2020), demonstrated promising accuracy for the identification of the origin of event candidates. We improve the performance of the classifier for LIGO-Virgo-KAGRA's fourth observing run by reviewing and updating the architecture and features used as inputs by the algorithm. We also retrain and fine-tune the classifier with data from the third observing run. To improve the prospect of electromagnetic follow-up observations, we incorporate GWSkyNet into LIGO-Virgo-KAGRA's low-latency infrastructure as an automatic pipeline for the evaluation of gravitational wave alerts in real time. We test the readiness of the algorithm on a LIGO-Virgo-KAGRA mock data challenge campaign. The results show that by thresholding on the GWSkyNet score, noise masquerading as astrophysical sources can be rejected efficiently and the majority of true astrophysical signals correctly identified.

What problem does this paper attempt to address?

The main problem that this paper attempts to solve is to improve the real - time classification accuracy of gravitational wave (GW) alerts in order to promote the success of electromagnetic (EM) follow - up observations. Specifically, the research team aims to better identify the real sources of gravitational wave events and reduce the false alarm rate by improving and updating the GWSkyNet machine - learning classifier. ### Background of the Main Problem 1. **Importance of Multi - Messenger Astronomy**: - Electromagnetic follow - up observations of gravitational wave events provide crucial scientific insights and help to gain a deeper understanding of these new astronomical transient sources. - Accurately identifying gravitational wave candidate events and quickly releasing sky - location information are essential for successful electromagnetic follow - up observations. 2. **Existing Challenges**: - There is a non - negligible false alarm rate when searching for gravitational wave candidate events in real - time. - A model that can reliably determine the sources of gravitational wave alerts is required to ensure the effectiveness of electromagnetic follow - up observations. ### Solution To address the above challenges, the research team developed and improved GWSkyNet, a machine - learning - based classifier. The main improvements include: - **Updates to Architecture and Input Features**: Re - evaluated and updated the input features and algorithm architecture used by the classifier to make it compatible with the alert data during O4 (the fourth observation run). - **Retraining and Fine - Tuning**: Retrained and fine - tuned the classifier using data from the third observation run (O3) to improve its performance. - **Low - Latency Infrastructure Integration**: Integrated GWSkyNet into the low - latency infrastructure of LIGO - Virgo - KAGRA as an automatic pipeline for real - time evaluation of gravitational wave alerts. ### Testing and Validation To verify the readiness and performance of GWSkyNet, the research team participated in the Mock Data Challenge (MDC) activity organized by LIGO - Virgo - KAGRA. The results showed that by setting an appropriate GWSkyNet scoring threshold, noise can be effectively rejected and most real astrophysical signals can be correctly identified. ### Conclusion Through these improvements, GWSkyNet shows higher accuracy and reliability in real - time classification of gravitational wave alerts, providing strong support for future electromagnetic follow - up observations.

GWSkyNet II : a refined machine learning pipeline for real-time classification of public gravitational wave alerts

GWSkyNet. II. A Refined Machine-learning Pipeline for Real-time Classification of Public Gravitational Wave Alerts

Explaining the GWSkyNet-Multi machine learning classifier predictions for gravitational-wave events

When to Point Your Telescopes: Gravitational Wave Trigger Classification for Real-Time Multi-Messenger Followup Observations

GSpyNetTree: A signal-vs-glitch classifier for gravitational-wave event candidates

Improving Early Detection of Gravitational Waves from Binary Neutron Stars Using CNNs and FPGAs

A machine-learning pipeline for real-time detection of gravitational waves from compact binary coalescences

Advancing Glitch Classification in Gravity Spy: Multi-view Fusion with Attention-based Machine Learning for Advanced LIGO's Fourth Observing Run

A new method to distinguish gravitational-wave signals from detector noise transients with Gravity Spy

Detection and Parameter Estimation of Gravitational Waves from Binary Neutron-Star Mergers in Real LIGO Data using Deep Learning

Gravitational wave search by time-scale-recursive denoising and matched filtering

Gravity Spy: Integrating Advanced LIGO Detector Characterization, Machine Learning, and Citizen Science

New Gravitational Wave Discoveries Enabled by Machine Learning

Localization of gravitational waves using machine learning

Convolutional Neural Networks for signal detection in real LIGO data

Improving Gravitational Wave Detection with 2D Convolutional Neural Networks

Deep Residual Networks for Gravitational Wave Detection

Rapid localization of gravitational wave sources from compact binary coalescences using deep learning

Demonstration of Machine Learning-assisted real-time noise regression in gravitational wave detectors