Jump to navigation Jump to content Jump to help Jump to language Jump to logout

505843-HS2025-0-Foundations of Deep Learning





Root number 505843
Semester HS2025
Type of course Lecture
Allocation to subject Statistics
Type of exam not defined
Title Foundations of Deep Learning
Description Deep learning has emerged as a powerful approach to solving complex problems in artificial intelligence, and understanding the underlying theory is crucial for practitioners and researchers alike. This postgraduate course offers an introduction to the theory behind deep learning, focusing specifically on mathematically rigorous results on the subject.
The course will start with an overview of the rudiments of statistical learning theory, such as loss functions, empirical risk minimization, kernel methods, generalization, and regularization. We will then thoroughly discuss the fundamentals of neural networks theory, covering topics such as architecture, activation functions, expressivity, approximation theorems, and training through (stochastic) gradient descent. The third part of the course will be devoted to some aspects of the optimization theory of neural networks. In particular, we will discuss the training dynamics of neural networks in the infinitely wide limit in two contrastive regimes: the neural tangent kernel regime and the mean-field regime. In the last part of th class we will provide an introduction to the fundamental theory of modern generative models such as transformers, diffusion models and flow matching techniques.
ILIAS-Link (Learning resource for course) Registrations are transmitted from CTS to ILIAS (no admission in ILIAS possible). ILIAS
Link to another web site
Lecturers Prof. Dr. Andrea Agazzi, IMSV - Gruppe Prof. Agazzi ✉
ECTS 6
Recognition as optional course possible Yes
Grading 1 to 6
 
Dates Tuesday 13:15-15:00 Weekly
Thursday 09:15-10:00 Weekly
 
Rooms Hörraum B001, Exakte Wissenschaften, ExWi
Hörraum B077, Exakte Wissenschaften, ExWi
 
Students please consult the detailed view for complete information on dates, rooms and planned podcasts.