Phantom dark energy as a natural selection of evolutionary processes $\hat{\rm a}$ $\textit{la}$ $\textit{genetic algorithm}$ and cosmological tensions

TL;DR

This paper uses machine learning to analyze cosmological data, supporting the phantom dark energy model as a way to reduce tensions in current cosmological measurements, and shows it fits data better than the standard model.

Contribution

It introduces a model-independent ML approach to compare cosmological models, revealing support for phantom dark energy and its effectiveness in alleviating tensions.

Findings

Supports phantom dark energy as a solution to cosmological tensions

Demonstrates better data fit than ΛCDM model

Reduces tensions in background and perturbative cosmological data

Abstract

We study the late-time cosmological tensions using the low-redshift background and redshift-space distortion data by employing a machine learning (ML) technique. By comparing the generated observables with the standard cosmological scenario, our findings indicate support for the phantom nature of dark energy, which ultimately leads to a reduction in the existing tensions. The model-independent approach also enables us to examine the combined background and perturbative history, where tensions are reduced. Moreover, from a statistical perspective, we have shown that our results exhibit a better fit to the data when compared to the $\Lambda$CDM model.