Overview

Graph representations are found in many areas of interest. In recent years, different Deep Learning models have been successfully applied to solve diverse tasks involving graph data.

In this seminar, the different components used in Grah-convolutional architectures will be presented, and literature about the topic will be suggested.

The students will afterward deepen their knowledge, by reviewing the literature on current open topics in research, and applications of Graph-Convolutional Networks. Finally, optional hands-on projects could be accomplished, granting additional CPs.

Course format

Every week, diverse material will be given (reading assignments, exercises...), and the material from last week will be reviewed.

- Meeting: Monday, 10:00-12:00, 2.70.0.08 Start 17.04.2023. If requested, the sessions will become hybrid and a zoom link will be provided.