AI and the World's Hard Problems · Physics · arXiv · 2026-10-22
Learning Parameters of dRGT Black Holes in Nonlinear Electrodynamics with Neural Network Scenarios
Machine learning is used to constrain parameters of AdS black holes in dRGT-like massive gravity coupled to nonlinear electrodynamics. The photon sphere, black hole shadow, and energy emission rate are computed with high-performance parallel computation, and a numerical framework constrains the electric charge and the nonlinear electrodynamic parameter using Event Horizon Telescope shadow observations. Neural networks then map the parameter regions consistent with EHT observations of M87*, SgrA*(VLTI), and SgrA*(Keck) within 1-sigma and 2-sigma confidence intervals, and the resulting constraints are found to be consistent with the SgrA*(VLTI) empirical data.