Structure of Quark Star: A Comparative Analysis of Bayesian Inference and Neural Network Based Modeling

TL;DR

This paper compares Bayesian inference and neural network models to analyze quark star structures, using observational data to constrain parameters and find consistent results supporting the conformal limit of the speed of sound.

Contribution

It introduces a comparative analysis of Bayesian and neural network methods for modeling quark star properties using multi-messenger astrophysical data.

Findings

Both methods yield consistent parameter estimates.

Predicted speed of sound aligns with the conformal limit.

Observational data effectively constrains quark matter models.

Abstract

In this work, we compare two powerful parameter estimation methods namely Bayesian inference and Neural Network based learning to study the quark matter equation of state with constant speed of sound parametrization and the structure of the quark stars within the two-family scenario. We use the mass and radius estimations from several X-ray sources and also the mass and tidal deformability measurements from gravitational wave events to constrain the parameters of our model. The results found from the two methods are consistent. The predicted speed of sound is compatible with the conformal limit.