Neural network-based gravitational wave interpolant with applications to low-latency analyses
Abstract
Matched-filter-based gravitational wave search pipelines identify candidate events within seconds of their arrival on Earth, offering a chance to guide electromagnetic follow-up and observe multimessenger events. Understanding the detectors’ response to an astrophysical transient across the searched signal manifold is paramount to inferring the parameters of the progenitor and deciding which candidates warrant telescope time. We describe a framework that uses artificial neural networks to interpolate gravitational waves and, equivalently, the signal-to-noise ratio across sufficiently local patches of the signal manifold. Our machine-learning based model generates a single waveform in 6 ms on a CPU and 0.4 ms on a graphical processing unit (GPU). When using a GPU to generate batches of waveforms simultaneously, we find that we can produce 10⁴ waveforms in ≲1 ms. This is achieved while remaining faithful, on average, to 1 part in 10⁴ (1 part in 10⁵) for binary black hole (binary neutron star) waveforms. The model we present is designed to directly utilize intermediate detection pipeline outputs in the hopes of facilitating a better real-time understanding of gravitational wave candidates.
Copyright and License
© 2024 American Physical Society.
Acknowledgement
The authors thank Jacob Golomb for Bilby assistance and useful comments. LIGO was constructed by the California Institute of Technology and Massachusetts Institute of Technology with funding from the National Science Foundation and operates under Cooperative Agreement No. PHY-1764464. This research has made use of data, software and/or web tools obtained from the Gravitational Wave Open Science Center, a service of LIGO Laboratory, the LIGO Scientific Collaboration and the Virgo Collaboration. Virgo is funded by the French Centre National de Recherche Scientifique (CNRS), the Italian Istituto Nazionale della Fisica Nucleare (INFN) and the Dutch Nikhef, with contributions by Polish and Hungarian institutes. This material is based upon work supported by NSF’s LIGO Laboratory which is a major facility fully funded by the National Science Foundation. The authors are grateful for computational resources provided by the LIGO Laboratory and supported by NSF Grants No. PHY-0757058 and No. PHY-0823459. This research has made use of data or software obtained from the Gravitational Wave Open Science Center (gwosc.org), a service of LIGO Laboratory, the LIGO Scientific Collaboration, the Virgo Collaboration, and KAGRA. This paper carries LIGO document number LIGO-P2400308.
Funding
The authors are grateful for computational resources provided by the LIGO Laboratory and supported by NSF Grants No. PHY-0757058 and No. PHY-0823459.
Files
Name | Size | Download all |
---|---|---|
md5:e8f27504cf577d9bcdb38d4e5d3c6537
|
1.0 MB | Preview Download |
Additional details
- Alternative title
- Accelerated signal-to-noise ratio interpolation across the gravitational-wave signal manifold powered by artificial neural networks
- National Science Foundation
- PHY-1764464
- National Science Foundation
- PHY-0757058
- National Science Foundation
- PHY-0823459
- Centre National de la Recherche Scientifique
- Istituto Nazionale di Fisica Nucleare
- Dutch Nikhef
- Accepted
-
2024-10-24Accepted
- Caltech groups
- LIGO
- Other Numbering System Name
- LIGO
- Other Numbering System Identifier
- LIGO-P2400308
- Publication Status
- Published