Scientific Machine Learning

Graduate / Professional

There are two main branches of technical computing: scientific computing and machine learning. Recently, there has been a convergence of the two disciplines in the emerging scientific machine learning (SciML) field. The main objective of this course is to teach theory, algorithms, and implementation of SciML techniques to graduate students. This course entails various methods to solve a broad range of computational problems frequently encountered in solid mechanics, fluid mechanics, nondestructive evaluation of materials, systems biology, chemistry, and non-linear dynamics. The topics in this course cover multi-fidelity learning, physics-informed neural networks, deep neural operators, uncertainty quantification, and parallel computing. Certain materials are discussed through student presentations of selected publications in the area. Students should have prior coursework in advanced calculus, linear algebra, and probability. Having a background in scientific computing, Python, and/or machine learning is helpful but not mandatory. Enrollment is limited; requires permission of the instructor. Th 4pm-5:55pm