[17a-S4_203-1~11] 13.1 Fundamental properties, surface and interface, and simulations of Si related materials

11:15 AM - 11:30 AM JST(2:15 AM - 2:30 AM UTC)

[17a-S4_203-9] Development of a Three-Dimensional Monte Carlo Device Simulator with Online Machine Learning for Enhanced Generality

〇(M1)Yusuke Otsuki¹, Nobuya Mori¹ (1.Osaka Univ.)

Keywords:

device simulation,machine learning

For predicting the performance of ultrascaled semiconductor devices, the development of three-dimensional device simulators that do not rely on empirical models is highly desirable. Monte Carlo (MC) device simulators are one of the promising candidates for this purpose. However, in conventional MC methods, the Poisson equation in a discretized space and the transport equations in continuous space must be solved iteratively, which leads to significant computational inefficiency. In this study, we develop a simulator that incorporates machine learning techniques, and we further introduce an online learning scheme to enhance its generality and adaptability.