Constraining Bianchi type-I cosmology in f(R,T) gravity with observational data: A Bayesian and ANN-based approach

This work investigates a Bianchi type-I cosmological model within the framework of [Formula: see text] gravity, adopting the functional form [Formula: see text]. We constrain the model parameters using a combination of observational datasets, including 55 points of Hubble parameter data, BAO measurements, and the Pantheon compilation of Type Ia Supernovae. Two methodologies are employed for parameter estimation: Bayesian inference via the Markov chain Monte Carlo (MCMC) method and an artificial neural network (ANN) approach implemented through the CoLFI framework. The study yields consistent estimates for key cosmological parameters such as [Formula: see text] and [Formula: see text], and determines the age of the Universe. In addition, we examine energy conditions, statefinder diagnostics and the Om[Formula: see text] parameter [Formula: see text] to assess the physical viability of the model. Our findings highlight the ANN’s effectiveness in cosmological inference, offering results comparable to MCMC with significantly reduced computational cost, and underscore the potential of machine learning in modern cosmological analysis.