MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data
Abstract
We present the results of \textbf{MADGRAV}, a deep-learning-based search for high-mass compact binary coalescences, applied to the data collected by the LIGO interferometers during the third observing run and during the first and second part of the fourth observing run. The \textbf{MADGRAV} pipeline consists of a series of sequential convolutional neural networks that perform anomaly detection, glitch classification, coherence testing, and signal ranking. Data from the Hanford and Livingston LIGO detectors are studied (both individually and in coherence) by way of 1 second Q-transform windows. Of the candidates that survive every stage of the pipeline, 48 reach the significance threshold, and we report 47 gravitational wave detections characterised by a false alarm rate below 1 yr−1 with a probability of astrophysical origin pastro>0.9. Of the 47 detections, 44 are shared with the minimally modelled coherent WaveBurst search. The observed total source-frame masses, extracted from official gravitational wave transient catalogues, are in the 14−236 M⊙ range with a median of 69 M⊙, and a median SNR of 16. We note that the recovered fraction of confident detections rises with mass: for LIGO detectors network SNR >10 the pipeline recovers 8.1% of confident catalog events below 30 M⊙, 39.8% between $30$ and 100 M⊙, and 53.3% above 100 M⊙, corresponding to 33.3%, 45.5% and 53.3% of the events detected by coherent WaveBurst in the same bins. These results suggest that anomaly detection pipelines can serve as an independent detection channel complementary to matched filtering in the high-mass high-SNR regime.
Related
- The Wavelet Detection Filter: a real time unmodeled pipeline for gravitational wave transients, ranking coincidences with a graph neural network
- LUMA: A CNN for Strong Gravitational Lens Searches in Astronomical Imaging
- A semicoherent resampling method for long-transient gravitational wave searches with applications to subsolar-mass primordial black-hole binaries
- Neural Network Guided Parameter Space Constraints for Gravitational Wave Searches from Binary Black Holes
- Alternative neural-network follow-up for all-sky FrequencyHough in continuous gravitational-wave searches
BibTeX
@misc{inguglia2026madgrav,
title = {{MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data}},
author = {Gianluca Inguglia and Huw Haigh and Ulyana Dupletsa and Alessandro Longo},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2609.39583},
eprint = {2609.39583},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.39583},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - MADGRAV: a multilevel anomaly-detection pipeline for gravitational-wave searches applied to LIGO data
AU - Gianluca Inguglia
AU - Huw Haigh
AU - Ulyana Dupletsa
AU - Alessandro Longo
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2609.39583
UR - https://arxiv.org/abs/2609.39583
AB - We present the results of \textbf{MADGRAV}, a deep-learning-based search for high-mass compact binary coalescences, applied to the data collected by the LIGO interferometers during the third observing run and during the first and second part of the fourth observing run. The \textbf{MADGRAV} pipeline consists of a series of sequential convolutional neural networks that perform anomaly detection, glitch classification, coherence testing, and signal ranking. Data from the Hanford and Livingston LIGO detectors are studied (both individually and in coherence) by way of 1 second Q-transform windows. Of the candidates that survive every stage of the pipeline, 48 reach the significance threshold, and we report 47 gravitational wave detections characterised by a false alarm rate below 1 yr^−1 with a probability of astrophysical origin p_astro>0.9. Of the 47 detections, 44 are shared with the minimally modelled coherent WaveBurst search. The observed total source-frame masses, extracted from official gravitational wave transient catalogues, are in the 14−236 M_⊙ range with a median of 69 M_⊙, and a median SNR of 16. We note that the recovered fraction of confident detections rises with mass: for LIGO detectors network SNR >10 the pipeline recovers 8.1% of confident catalog events below 30 M_⊙, 39.8% between $30$ and 100 M_⊙, and 53.3% above 100 M_⊙, corresponding to 33.3%, 45.5% and 53.3% of the events detected by coherent WaveBurst in the same bins. These results suggest that anomaly detection pipelines can serve as an independent detection channel complementary to matched filtering in the high-mass high-SNR regime.
ER -