Convolutional neural network for retrieval of the time-dependent bond length in a molecule from photoelectron momentum distributions

Abstract

We apply deep learning for retrieval of the time-dependent bond length in the dissociating two-dimensional H+2 molecule using photoelectron momentum distributions. We consider a pump-probe scheme and calculate electron momentum distributions from strong-field ionization by treating the motion of the nuclei classically, semiclassically or quantum mechanically. A convolutional neural network trained on momentum distributions obtained at fixed internuclear distances retrieves the time-varying bond length with an absolute error of 0.2–0.3 a.u.