Scientific Machine Learning: Mathematical Foundations and Applications

Catalog description

This course is an introduction to scientific machine learning, which is a rapidly developing field at the intersection of scientific computing, machine learning (ML). In contrast with more standard applications of ML, scientific ML relies on modern numerical analysis to provide rigorous guarantees. This course will cover several numerical methods and study their mathematical foundations. It will include the following topics: (1) neural networks as function approximators, (2) deep learning methods for partial differential equations and (3) machine learning methods for optimal control problems. The exact list of topics may vary each year to reflect the development of this field. Prerequisite: MATH-SHU 140 Linear Algebra and CSCI-SHU 11 Introduction to Computer Programming Fulfillment: Math Additional Math elective; Honors Math Math elective.