- Start Date:
June 5, 2012
- Event Start Time:
9:30 AM
- Event End Time:
10:30 AM
- Organizers:
James Abello
- Seminar Series:
REU Seminar
- Presenter(s):
Milan Bradonjic - Bell Labs, Alcatel-Lucent
- Event Location:
DIMACS Seminar room
- Abstract:
One of the transitions from random graph models given by Erdos and Renyi to the models that may describe real processes in a more realistic manner is the model of Random Geometric Graphs (RGGs). An RGG is a graph resulting from placing nodes independently and uniformly at random on the unit square, and creating edges between pairs of nodes if their Euclidean distance is at most some given value. There has been a flurry of work on RGGs in the last two decades. The RGG model has been used in disciplines such as: wireless sensor networks, cluster analysis, statistical physics, hypothesis testing, and modeling data in a high-dimensional space. The first session will introduce the necessary background and the main conceptual tools for this line of research. Seminal results on the connectedness and phase transition in RGGs will be discussed. The second session will present recent results on RGGs, as well as introduce a new structured random generative model of Geographical Threshold Graphs. Open avenues of research in this area will be discussed during the second session and software tools for further experimentation will be introduced.
