Bavarian AI Chair for Mathematical Foundations of Artificial Intelligence

Breadcrumb Navigation


Mathematical and Statistical Foundations of Machine Learning

Welcome to the lecture "Mathematical and Statistical Foundations of Machine Learning"

by Prof. Dr. Gitta Kutyniok.

Course Description

Machine learning has led to spectacular success in many applications, especially after the development of deep learning. This course aims to provide an introduction into selected topics of machine learning and statistical learning theory. We will start with common linear models in machine learning (e.g. perceptron, Adaline, support vector machines) and proceed to non-linear models, such as kernel methods. At the end of the course, we will briefly touch upon theoretical foundations of multi-layered neural networks and deep learning as well, also as preparation for a follow-up course in the Winter Term 2022/23.

The minimum goal is to arrive at a profound understanding of both the mathematical and statistical foundations of machine learning as well as the implementation of common machine learning methods in Python

Schedule and Venue

Lecture: Monday 14-16 (Room B005) Prof. Kutyniok
Wednesday 14-16 (Room B006)
Exercise Class: Tuesday 10-12 (Room B005) Mariia Seleznova, Adalbert Fono
Monday 10-12 (via Zoom)
Tutorials: Tuesday 16-18 (B139)
Wednesday 16-18 (via Zoom)
Thursday 12-14 (B041)



Course Materials

Course materials will be provided via the uni2work framework. Please register for our course on uni2work.


This course is offered mainly to master students in Mathematics, Financial Mathematics, Theoretical and Mathematical Physics, Data Science and Statistics. Bachelor students in Mathematics and Financial Mathematics may also attend the course and take the exam. However, the course can not be credited as a bachelor module. Upon passing the exam, the course can only be credited (as described below) later in a subsequent master program in Mathematics or Financial Mathematics.

We require good knowledge of Probability Theory and Linear Algebra. A basic knowledge in Python or a similar programming language and access to a computer with a Python development environment is required in order to complete the tutorials.

Creditable Modules

  • Master in Mathematics: WP5 or WP35 in PO2021, WP5 in PO2011
  • Master in Financial and Insurance Mathematics: WP20 or WP22 in PO2021, WP21 or WP23 in PO2019