Physics-Informed Machine Learning

Physics-Informed Machine Learning SS 2025

Time and Place

Offered every Summer Semester
Lecture: Thursdays 8⁰⁰-9³⁰
Exercise: Thursdays 12⁰⁰-12⁴⁵
Place: 5510.01.050 & online via Zoom.us (for meetingID & password check Moodle)

Program

MSc.
Language: English
5 ECTS, 2 VO + 1 UE

Links

Module Handbook
Course Calendar
Moodle

Literature

Bishop: Pattern Recognition and Machine Learning, Theodoridis: Machine Learning. Extra literature will be given in the lecture notes.

Exam

Type: written
Time & Place: TBA
Allowed: non-programable calculator

Exam Breakdown

The exam grade can be increased by 0.3 if exercises are submitted and sufficient attempts are made to solve them.

What's this course about?

In this course, you will get to know some of the widely used machine learning techniques. We will cover methods for classification and regression, methods for clustering and dimensionality reduction, and generative models. In the exercise class, you will transform the theoretical knowledge into practical knowledge and learn how to use the machine learning tools to solve engineering problems. For a more detailed description see the module handbook.

How does it work?

Learn the basics

The theoretical background will be given in 2h/week lectures with a mixture of slides (motivational examples, key concepts), blackboard (important mathematical background), and animations (algorithm demonstrations).

Apply to Problems

Exercise handouts are given beforehand. During exercise class, TAs give you hints on how to solve them and explain any questions about the previous exercise. Computational problems can be solved in any programming language. However, the solutions will be in Python.