SiGMa-Net: Deep learning network to distinguish binary black hole signals from short-duration noise transients

Abstract

Blip glitches, a type of short-duration noise transient in the LIGO--Virgo data, are a nuisance for the binary black hole (BBH) searches. They affect the BBH search sensitivity significantly because their time-domain morphologies are very similar, and that creates difficulty in vetoing them. In this work, we construct a deep-learning neural network to efficiently distinguish BBH signals from blip glitches. We introduce sine-Gaussian projection (SGP) maps, which are projections of GW frequency-domain data snippets on a basis of sine-Gaussians defined by the quality factor and central frequency. We feed the SGP maps to our deep-learning neural network, which classifies the BBH signals and blips. Whereas the BBH signals are simulated, the blips used are taken from real data throughout our analysis. We show that our network significantly improves the identification of the BBH signals in comparison to the results obtained using traditional-${\chi }^2$ and sine-Gaussian ${\chi }^2$. For example, our network improves the sensitivity by 75% at a false-positive rate of $10^{-2}$ for BBHs with total mass in the range $[80,140]M_{\odot }$ and SNR in the range $[3,8]$. Also, it correctly identifies 95% of the real GW events in GWTC-3. The computation time for classification is a few minutes for thousands of SGP maps on a single core. With further optimisation in the next version of our algorithm, we expect a further reduction in the computational cost. Our proposed method can potentially improve the veto process in the LIGO--Virgo GW data analysis and conceivably support identifying GW signals in low-latency pipelines.

11 pages, 8 figures and 2 tables. Reviewed by LIGO Scientific Collaboration (LSC) with preprint number LIGO-P2100485

References in corpus (11)

• Deep Learning in Neural Networks: An Overview
• Advanced LIGO
• Gravity Spy: Integrating Advanced LIGO Detector Characterization, Machine Learning, and Citizen Science
• LIGO Detector Characterization in the Second and Third Observing Runs
• The Science Case for LIGO-India
• Discovering features in gravitational-wave data through detector characterization, citizen science and machine learning
• Detection of gravitational-wave signals from binary neutron star mergers using machine learning
• Generalized Approach to Matched Filtering using Neural Networks
• Accelerating Recurrent Neural Networks for Gravitational Wave Experiments
• A classifier for gravitational-wave inspiral signals in non-ideal single-detector data
• A unified approach to ${\chi }^2$ discriminators for searches of gravitational waves from compact binary coalescences

Cited by in corpus (9)

• Deep Residual Networks for Gravitational Wave Detection
• Gravity Spy: Lessons Learned and a Path Forward
• New Gravitational Wave Discoveries Enabled by Machine Learning
• Applications of machine learning in gravitational wave research with current interferometric detectors
• Premerger detection of massive black hole binaries using deep learning
• SLICK: Strong Lensing Identification of Candidates Kindred in gravitational wave data
• Neural network time-series classifiers for gravitational-wave searches in single-detector periods
• Novel Deep Learning Approach to Detecting Binary Black Hole Mergers
• Improved binary black hole searches through better discrimination against noise transients