Phantom dark energy as a natural selection of evolutionary processes $\widehat{\mathrm{a}}$ $\mathit{\text{la}}$ $\mathit{\text{genetic}}\mathit{\text{algorithm}}$ and cosmological tensions

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 $\mathrm{\Lambda }$CDM model.