[9a-S301-1~7] 23.1 Joint Session N "Informatics"

11:00 AM - 11:15 AM JST(2:00 AM - 2:15 AM UTC)

[9a-S301-6] Total energy prediction of GeSn alloy semiconductors using extreme learning machine

〇Yusuke Noda¹, Hibiki Bekku², Koji Sueoka³ (1.Kyushu Inst. Technol., 2.Grad. Sch. Comput. Sci. Syst. Eng., Okayama Pref. Univ., 3.Fac. Comput. Sci. Syst. Eng., Okayama Pref. Univ.)

Keywords:

IV-group mixed alloy,first-principles calculation,machine learning

In this study, a machine learning model was constructed to predict the total energy of crystal structures of GeSn alloy semiconductors using the Extreme Learning Machine (ELM), which enables rapid construction of regression models. The predictive accuracy of the ELM regression model was evaluated after it learned the relationship between the total energy of Ge_1-xSn_x alloys (0.0 < x < 1.0) obtained from DFT calculations and descriptors representing the atomic environment.