Inhalt des Dokuments
Seminar on Network Evolution Models
Networks are dynamic objects in constant evolution: new vertices appear that need to be connected via the creation of additional edges, while unused links are sometimes removed. Many works have proposed and studied models to understand this evolution, starting with the model of preferential attachment, which assumes that new nodes connect to existing nodes with a probability proportional to their degree. Decentralized decision making has also been studied in various models for the evolution of the Internet or social networks in the algorithmic game theory literature. However, these models do not seem to be well suited to the study of transportation networks, which are embedded in a two-dimensional space, and evolution models for spatial networks are still at a rudimentary stage.
In this seminar, we will review old and recent approaches to model and analyze the evolution of social networks, communication networks, and transportation networks, from various points of view. The goal is to make students familiar with developments in this area, and to teach them the necessary tools to explore scientific papers and literature on their own. Moreover, students will acquire important skills for presenting these technical results to a wider audience.
The list of papers, important dates and the zoom-links for the meetings for this seminar are available on ISIS .
The purpose of this seminar is to read and understand recent results from the literature on graphs, algorithms and optimization. Participants are expected to have some experience with the analysis of advanced algorithms (e.g. from an ADM course).