We will focus on programming; however, the necessary mathematical background will be covered in the math lab as needed.

Course materials will be based on the book Dive into Deep Learning (https://d2l.ai/), and topics will include:

• Multilayer perceptrons
• Modern convolutional neural networks
• Recurrent neural networks
• Attention mechanisms
• Optimization algorithms
• Gaussian Process
• Hyperparameter optimization

The course may integrate practical projects developed by participating PhD and Master’s students. This includes hands-on activities such as running code from GitHub, performing inference, and, where feasible, experimenting with model training.

Prerequisites: Python programming, basic knowledge of linear algebra, calculus and probability theory are highly desirable. However, we will try to make the course as self-contained as possible.

You need an account at University of Greifswald. The course will be in English.

Instructor: Alexandra Moringen, alexandra.moringenuni-greifswaldde
First meeting: Thursday, 23. October, 4pm
Dates: Winter Semester 2025/26
Location: Sensorlab, 0.20, Universitätsrechenzentrum
Available seats: 12

Equipment needed:
A laptop with a browser and Internet access for programming.

Please register for this workshop before the course starts via our online course registration.

Contact
Graduiertenakademie der Universität Greifswald
Domstraße 11, Eingang 4, Raum 3.24, 17489 Greifswald
Telefon +49 3834 420 1618
graduiertenakademieuni-greifswaldde