AI-Based Prediction of Electronic Properties of GaAs Materials

Abstract

Advancements of machine learning algorithms in analysing the atomistic structure and properties of quantum confined nanostructures are considered by all kinds of techniques. In this work, a regression based fine tree machine learning algorithm is applied to solve the current-voltage characteristics model of GaAs nanotube under quantum confinement effect. The length of the nanotube is 3.52 nm and the width is 3.61 nm. In this paper, we estimate the predictive distribution of the models of current-voltage characteristics with enough confidence level. This is not a simple task as the quantum confined nanostructure has a backscattering effect due to the mean-free path of the channel length. With such kind of quantum interference, there are challenges in predicting the current voltage characteristics. Hence, with the help of this machine learning algorithm, we have estimated this current-voltage characteristics model with negligible error rate.