Real-time gravitational-wave science with neural posterior estimation

TL;DR

This paper introduces DINGO, a neural network-based method for rapid and accurate gravitational-wave parameter estimation, achieving near real-time inference with results comparable to traditional methods.

Contribution

The paper presents a novel deep learning approach that significantly reduces inference time for gravitational-wave data analysis while maintaining high accuracy, trained on simulated data including noise characteristics.

Findings

Inference time reduced from days to minutes per event

High agreement with standard Bayesian inference results

Enables real-time gravitational-wave data analysis

Abstract

We demonstrate unprecedented accuracy for rapid gravitational-wave parameter estimation with deep learning. Using neural networks as surrogates for Bayesian posterior distributions, we analyze eight gravitational-wave events from the first LIGO-Virgo Gravitational-Wave Transient Catalog and find very close quantitative agreement with standard inference codes, but with inference times reduced from O(day) to a minute per event. Our networks are trained using simulated data, including an estimate of the detector-noise characteristics near the event. This encodes the signal and noise models within millions of neural-network parameters, and enables inference for any observed data consistent with the training distribution, accounting for noise nonstationarity from event to event. Our algorithm -- called "DINGO" -- sets a new standard in fast-and-accurate inference of physical parameters of detected gravitational-wave events, which should enable real-time data analysis without sacrificing accuracy.