Bedartha Goswami


20 March 2023 – 24 March 2023
Monday–Friday, 09:00 – 16:00
10 lectures of 1.5 hours each, 10 tutorials of 1.5 hours each


Seminar Room 4F03, Geo- und Umweltforschungszentrum (GUZ), Morgenstelle

Credits and Workload

Workload: 60 h
Contact hours: 40 h
Self study: 20 h
Duration: 1 week Medium of instruction: English


Presentation of recap in class; Written final assignment

About the course

The course will cover the following topics related to non-neural-network-based machine learning:

  • Neural networks for tabular data, images, and sequences
  • Autoencoders
  • Representation Learning
  • Graph embeddings
  • Regularisation, Optimisation, Approximate Inference

Teaching methods

The lectures are designed to be interactive and dialogic. Lectures are going to be conducted with a black board (no slide presentation). Tutorials will be conducted with a laptop.

Learning target

At the end of the course, students will posses a basic understanding of what is deep learning and what are the basic tools and techniques that it can offer us to model patterns in complex data by learning smart and efficient representaions.


machine learning, statistics, complex networks, clustering, regression

Enrollment information

Enrollment is via ALMA

Tentative lecture plan