Date & Time:

   2026 / 11 / 06 (Fri) 14:20 - 16:20

Location:

   Delta Building R216, NTHU

Speaker:

   薛名成 教授

   國立陽明交通大學 應用數學系

Topic:

   From Mathematical Modeling to Scientific Machine Learning: Classical Numerical Methods and PINNs

Abstract:

   Mathematical modeling and numerical methods are essential tools for understanding and predicting real-world phenomena in science and engineering. In this lecture, we will introduce three classical approaches for solving partial differential equations: the finite difference method (FDM), finite volume method (FVM), and finite element method (FEM). Meanwhile, we will then explore physics-informed neural networks (PINNs), which integrate physical laws into machine learning. Finally, we will discuss the connections, advantages, and challenges of these approaches, highlighting the evolving role of mathematics in scientific computing.

Short Bio:

   Ming-Cheng Shiue is a Professor in the Department of Applied Mathematics at National Yang Ming Chiao Tung University, Taiwan. His research interests include numerical analysis, scientific computing, partial differential equations, and geophysical fluid dynamics. His work focuses on the development and mathematical analysis of numerical methods for fluid dynamics and related scientific problems, with recent interests in scientific machine learning, particularly physics-informed neural networks (PINNs).