Cosmological Inference using Gravitational Waves and Normalising Flows

TL;DR

This paper introduces a machine learning method using normalising flows to efficiently infer cosmological parameters, like the Hubble constant, from gravitational wave data, incorporating galaxy catalogues and prior information.

Contribution

The paper presents a novel, fast, and generalizable normalising flow approach for cosmological inference from gravitational wave events, applicable to various compact binary coalescences and cosmological models.

Findings

Achieved a rapid estimate of the Hubble constant in about 1 second.

Provided a Bayesian posterior estimate of H0 with confidence bounds.

Demonstrated the method on real gravitational wave data from LIGO/VIRGO.

Abstract

We present a machine learning approach using normalising flows for inferring cosmological parameters from gravitational wave events. Our methodology is general to any type of compact binary coalescence event and cosmological model and relies on the generation of training data representing distributions of gravitational wave event parameters. These parameters are conditional on the underlying cosmology and incorporate prior information from galaxy catalogues. We provide an example analysis inferring the Hubble constant using binary black holes detected during the O1, O2, and O3 observational runs conducted by the advanced LIGO/VIRGO gravitational wave detectors. We obtain a Bayesian posterior on the Hubble constant from which we derive an estimate and 1$\sigma$ confidence bounds of $H_{0} = 74.51^{+14.80}_{-13.63} \: \text{km} \:\text{s}^{-1} \text{Mpc}^{-1}$. We are able to compute this result in $\mathcal{O}(1)$ s using our trained Normalising Flow model.