A neural network-based gravitational wave interpolant with applications to low-latency analyses

TL;DR

This paper introduces a neural network-based interpolant for gravitational wave signals that enables rapid waveform generation, significantly improving low-latency analysis for real-time gravitational-wave detection and follow-up observations.

Contribution

It presents a novel neural network framework that interpolates gravitational wave signals across the parameter space, enabling ultra-fast waveform generation for real-time analysis.

Findings

Waveform generation in 6 ms on CPU and 0.4 ms on GPU.

Batch processing of 10,000 waveforms in under 1 ms on GPU.

High fidelity with 1 part in 10^4 for black hole binaries and 1 part in 10^5 for neutron star binaries.

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 multi-messenger 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 (SNR) across sufficiently local patches of the signal manifold. Our machine-learning based model generates a single waveform in 6 milliseconds on a CPU and 0.4 milliseconds on a GPU. When using a GPU to generate batches of waveforms simultaneously, we find that we can produce $10^4$ waveforms in $\lesssim 1$ ms. This is achieved while remaining faithful, on average, to 1 part in $10^4$ (1 part in $10^5$) 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.