Scientific machine learning

Summary

This course covers topics of scientific machine learning. Part I focuses on sequential-in-time training of nonlinear parametrizations for numerically solving partial differential equations. Part II treats generative modeling of physical processes. This is a project-based course.

Content

Learning Prerequisites

Required courses

Math-250: Advanced Numerical Analysis I

Math-351: Advanced Numerical Analysis II

Math-414: Stochastic simulation

Recommended courses

Students should be comfortable with Python programming.

Learning Outcomes

By the end of the course, the student must be able to:

• Apply scientific machine learning techniques
• Explain the main concepts and methods of scientific machine learning
• Assess / Evaluate scientific machine learning techniques in terms of scope, accuracy, and computational costs

Teaching methods

- Lectures

- Exercises

Expected student activities

Attending lectures
Implementing mathematical methods in a programming language
Finishing exercises

Assessment methods

40% homeworks and projects and 60% final exam

Resources

Moodle Link

• https://go.epfl.ch/MATH-507