Domain-optimized machine-learning tool for gravitational-wave inference
Physical Review D · 28.07.2026 · 114 (2)
DOI: 10.1103/2l59-vly4 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physical Review D, online erschienen am 28.07.2026. Verfasst von Javier Roulet und 3 weiteren Autor:innen. Die Arbeit nennt 122 Literaturangaben. Gefördert von: Sherman Fairchild Foundation, European Commission, H2020 Marie Sk?odowska-Curie Actions.
Abstract (Originalsprache)
Fast and reliable inference of gravitational-wave source parameters is crucial for analyzing large catalogs that are reaching the size of hundreds of detections, and for identifying short-lived electromagnetic counterparts. Neural posterior estimation has emerged as a powerful inference method, where the model is trained on simulated gravitational-wave data at considerable computational cost, but thereafter enables extremely fast and inexpensive inference at test time. Here, we extend this approach by incorporating domain-specific physical insights and methods in the model architecture. These include compressing the data by heterodyning against a reference waveform chosen via approximate likelihood maximization, removing parameter degeneracies through tailored coordinate systems, and eliminating known multimodalities by folding the parameter space. As a result, the network is approximately equivariant to changes in the source parameters, and achieves a reduced training cost and improved model interpretability. Our implementation, called labrador, can be trained end-to-end on a 1-day timescale on $\sim 10^2$ CPU cores and a V100 GPU, achieving a median importance-sampling efficiency of 1% on quadrupolar, aligned-spin signals in a broad mass range (chirp mass $\mathcal{M} \in 1\text{-}50\,\mathrm{M}_\odot$, mass ratio $q>0.1$). labrador is the first neural inference code to achieve extensive coverage of long-duration signals with secondary masses $m_2
Förderung
- Sherman Fairchild Foundation
- European Commission
- H2020 Marie Sk?odowska-Curie Actions
- National Science Foundation
- Science and Technology Facilities Council
- Max Planck School of Photonics
- Australian Research Council
- Centre National de la Recherche Scientifique
Zitieren
APA
Roulet, J., Crisostomi, M., Thomas, L. M., & Chatziioannou, K. (2026). Domain-optimized machine-learning tool for gravitational-wave inference. Physical Review D, 114(2). https://doi.org/10.1103/2l59-vly4
Vancouver
Roulet J, Crisostomi M, Thomas LM, Chatziioannou K. Domain-optimized machine-learning tool for gravitational-wave inference. Physical Review D. 2026;114(2). doi:10.1103/2l59-vly4
BibTeX
@article{Roulet2026,
author = {Roulet, Javier and Crisostomi, Marco and Thomas, Lucy M. and Chatziioannou, Katerina},
title = {Domain-optimized machine-learning tool for gravitational-wave inference},
journal = {Physical Review D},
year = {2026},
volume = {114},
number = {2},
doi = {10.1103/2l59-vly4}
}
EUVIMED