Reconciling early and late time tensions with Reinforcement Learning

We study the possibility of accommodating both early and late-time tensions using a novel reinforcement learning technique. By applying this technique, we aim to optimize the evolution of the Hubble parameter from recombination to the present epoch, addressing both tensions simultaneously. To maximize the goodness of fit, our learning technique achieves a fit that surpasses even the ΛCDM model. Our results demonstrate a tendency to weaken both early and late time tensions in a completely model-independent manner.