Usage of machine-learning algorithms in inverse problem of light self-focusing in isotropic chiral medium with cubic nonlinearity

Abstract

Abstract Efficiency of convolutional artificial neural networks in the problem of finding nonlinearity parameters proportional to the local and non-local cubic dielectric susceptibilities of a medium and intensity of the incident radiation, fully describing the self-focusing character of elliptically polarized laser beams, is demonstrated. It is shown that realization of the predictive algorithm by the neural network can be improved by using complex structured light so that the error is lowered down to percent units.