Constraining Ricci-Cubic Holographic Dark Energy from observational data using the MCMC sampling and enhanced Machine learning analysis

Abstract: In this work, we find constraints on the parameter space of the Ricci-Cubic Holographic dark energy (RCHDE) from various observational data sets like Hubble data, cosmic-chronometer data, Baryon-acoustic oscillation data, and also data from gamma-ray bursts. RCHDE is formed from the cubic invariant, which in turn is built from the cubic contractions of the Riemann and Ricci tensors. We have used the Markov chain Monte-Carlo (MCMC) sampling technique to find constraints on the model parameters via Bayesian inference. Contour plots have been obtained for the model parameters, showing their marginalized and joint probability distributions. The best-fit regression lines are found for the constrained model and compared with the standard $\Lambda$CDM model to verify and validate the model. To complement this data analysis mechanism, we have also performed an enhanced machine learning analysis using observational Hubble parameter data. This approach serves to validate the model's predictive power through independent, data-driven regression techniques. Different graphical illustrations of the machine learning techniques have been presented to understand the results. These illustrations reveal a strong agreement between the Hubble parameter predictions from the machine learning models, the theoretical RCHDE model, and observational data.