Title: Caterina De Bacco, Cyber Valley
Start date: Mar 26 2019
Start time: 03:00 pm
End time: 04:00 pm
Location: MRZ seminar room

Abstract: Networks are a standard tool to represent interactions between individual elements in a system. Given the widespread nature of such entities, in recent years many network-based models have been proposed to study a range of problems that apply to various disciplines. Among them, addressing inference tasks have been a popular research direction to enable the characterization and understanding of the main properties that drive these types of systems.
The goal of this talk is to give an overview of two such inference tasks that have been particularly successful in tackling interdisciplinary applications: the community detection problem and inferring rankings from pairwise comparisons. The plan is to show the main model assumptions and methodological approaches to tackle each of them with the goal in mind to raising awareness of how they work and possibly stimulate their use in disciplines that are less familiar with them.

Host: Anna Levina