Graph Machine Learning

Graph data are present in a myriad of modern computer sciences systems and applications. Examples include data generated over social, brain, financial, power, water and sensor networks. Because these data have a complicated structure they require different tools conventionally used in machine and deep learning to develop end to end solutions. These solutions falls generally under the umbrella of graph-based machine learning and can be used to perform recommendations, detect anomalies in the brain, predict financial crisis, estimate the sate of a power or water network, and coordinate group of autonomous moving sensor, to name a few.

This course deals with the foundations and principles of machine and deep learning for network data. Topics include: unsupervised and semi-supervised learning on graphs; graph representation learning; graph signal processing; graph convolutions; graph neural networks; spatiotemporal learning on graphs; scalable algorithms; explainability and privacy of graph neural networks.

The final grade of the course consists of the following components:

• Oral Exam: Individual oral exam (weighting 40%)

• Group Report: Project report (weighting 60%)

Final grade calculation = 0.4 * Oral Exam + 0.6 * Group Report

A passing final grade for the course can only be earned when for all components at least a 5.0 is earned, and the weighted final grade is at least a 5.8.

In case of an insufficient final result, repair options may exist in accordance with Article 17A, Times and number of examinations, sub 1, of the Teaching and Examination Regulations, for:

• Oral Exam: Resit opportunity

• Group Report: Repair opportunity

Disclaimer: information may change depending on unforeseen circumstances or measures (see: TER Art 2, sub 5).