Présentation sur YouTube.

Molecular geometry optimization plays a central role in computational chemistry, relying heavily on the efficient computation of interatomic distances and their derivatives. This talk presents a novel geometric framework based on a non-Euclidean model of the 3D space, designed to improve the performance of energy minimization algorithms by simplifying distance and derivative calculations.