AI-Assisted Detection of Residual Time-Field (γ) in Hilltop Inflationary Cosmology

This research integrates AI with theoretical cosmology to investigate residual Time-field (γ) effects post-Hilltop inflation. The study combines a deep neural network surrogate model with Bayesian MCMC inference using Planck PR4 and Pantheon+ datasets. Key results indicate a residual energy density γ ≈ 0.0015, which partially alleviates the Hubble tension. The workflow is fully reproducible and includes defense-ready visualization (H(z) deviation, likelihood contours, project progress chart).