Fast Radio Bursts and Artificial Neural Networks: a cosmological-model-independent estimation of the Hubble Constant

TL;DR

This paper presents a novel neural network-based method to estimate the Hubble Constant from Fast Radio Bursts data, providing a model-independent approach that could improve precision with future observations.

Contribution

It introduces a new neural network method for cosmological parameter estimation using FRBs, independent of specific cosmological models.

Findings

Estimated Hubble constant as 67.3±6.6 km/s/Mpc from current data

Achieved ~10% precision with 23 localised FRBs

Forecasted future precision comparable to leading cosmological measurements

Abstract

Fast Radio Bursts (FRBs) have emerged as powerful cosmological probes in recent years offering valuable insights into cosmic expansion. These predominantly extragalactic transients encode information on the expansion of the Universe through their dispersion measure, reflecting interactions with the intervening medium along the line of sight. In this study, we introduce a novel method for reconstructing the late-time cosmic expansion rate and estimating the Hubble constant, solely derived from FRBs measurements coupled with their redshift information while employing Artificial Neural Networks. Our approach yields a Hubble constant estimate of $H_0 = 67.3\pm6.6\rm \ km \ s^{-1} \ Mpc^{-1}$. With a dataset comprising 23 localised data points, we demonstrate a precision of $\sim10\%$. However, our forecasts using simulated datasets indicate that in the future it could be possible to achieve precision comparable to the SH0ES collaboration or the Planck satellite. Our findings underscore the potential of FRBs as alternative, independent tools for probing cosmic dynamics.