|
Instructors: Univ.-Prof. Dr. Michael Wand
Event type:
Lecture/practice class
Displayed in timetable as:
08.079.318
Hours per week:
4
Credits:
6,0
Language of instruction:
German
Min. | Max. participants:
- | -
Requirements / organisational issues:
The lecture builds on top of "Modelling 1" but a good math background (from minor or major studies in math or physics) is usually also sufficient.
Contents:
The lecture modelling 2 aims at a deeper understanding of current "deep learning" methods. It will start with an overview of classical machine learning topics (with a focus on the theoretical of generalization), provide some in-depth information on topics complementary to the separate "machine learning" lecture, and in the last part, will go into details of how to understand the perculiar behavior of deep networks (e.g. double descent phenomena, neural tangent kernels, fischer information, phase-transitions).
Additional information:
Important: Please make sure to visit
https://luna.informatik.uni-mainz.de/mod2-23/
and sign up for our mattermost channel (see web site for details).
|
