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

We employ a machine learning (ML) algorithm to analyze cosmological background data and the linear red-shift space distortion (RSD) data in a model-independent way, with specific focus on the Hubble expansion rate and the growth of large-scale structure. We find strong evidence that the natural enhancement in the Hubble parameter at low redshifts is due to the underlying phantom nature of dark energy, rather than low matter density. As for the RSD data, we find a higher value of $\sigma^8_{(0)}$ which is consistent with CMB's predictions, but the outcome of low matter density leads to unresolved tension. This might point towards a new physical phenomenon at the perturbative level in the low redshift regime. From a statistical perspective, we have demonstrated that our results hold greater preference compared to those obtained by the $\Lambda$CDM model.