[13p-PA4-1~3] 13.1 Fundamental properties, surface and interface, and simulations of Si related materials

[13p-PA4-2] Modeling of electron transmission and time-development of electron density distribution
in the nanoscale device under random impurity fluctuation with machine learning

Ryuho Nakaya¹, Souma Kawahara¹, Yoshitaka Itoh¹, Tota Suko², 〇Masakazu Muraguchi¹ (1.Hokkaido Univ. of Science, 2.Waseda Univ.)

Keywords:Electron dynamics, Machine Learning

We report the application of machine learning to the time evolution calculation of electron wave functions. Based on the results of the electronic dynamics calculations for semiconductors, a model was created using random forest (RF) and neural network (NN) that can predict electron transmission when an impurity distribution is given as an input. From the comparison of the two, we report the results of creating a more accurate model and the modeling of electron density over time by learning image sequences using NN.