Aims of the course

Networks are a powerful tool to represent datasets with pairwise interactions between individual units. Relevant examples are social networks, representing interactions between people, biological datasets as protein-protein interactions or gene-disease associations; financial transactions between institutions or companies. In this course, we will learn how to perform inference tasks in networked datasets, where the minimal input information is a list of edges. We will cover how to learn hidden patterns like community structure (or node clustering) or hidden hierarchies of nodes. We will use probabilistic methods and inference techniques like expectation-maximization and variational inference.

Content(sketch):

(1-2) Stochastic block model and community detection;
(3-4) Variational inference on networks;
(5) Ranking from pairwise comparisons;
(6) Non-linear models: deep learning architectures for networked datasets.

Schedule:

• Monday 05/05/2025 12:30-14:30 – Room LuF1
• Wednesday 07/05/2025 12:30-14:30 – Room 1A150
• Thursday 08/05/2025 16:30-18:30 – Room 1A150
• Monday 12/05/2025 12:30-14:30 – Room LuF1
• Tuesday 13/05/2025 16:30-18:30 – Room 1A150
• Thursday 15/05/2025 16:30-18:30 – Room 1A150