Indexed on: 17 Feb '17Published on: 17 Feb '17Published in: The Journal of chemical physics
A machine learning approach based on the artificial neural network (ANN) is applied for the configuration problem in solids. The proposed method provides a direct mapping from configuration vectors to energies. The benchmark conducted for the M1 phase of Mo-V-Te-Nb oxide showed that only a fraction of configurations needs to be calculated, thus the computational burden significantly decreased, by a factor of 20-50, with R(2) = 0.96 and MAD = 0.12 eV. It is shown that ANN can also handle the effects of geometry relaxation when properly trained, resulting in R(2) = 0.95 and MAD = 0.13 eV.