ECE seminar series: Dynamical Processes on ‘Coupled’ Complex Networks: Cascading Failures, Information Epidemics, and Complex Contagions
Wednesday, March 6, 2013 – ( 3:00 pm to 4:00 pm )
Location: Babbio Center, Room 319
Dynamical Processes on ‘Coupled’ Complex Networks: Cascading Failures, Information Epidemics, and Complex Contagions
BY Dr. Osman Yağan
CyLab, Carnegie Mellon University
Pittsburgh, PA 15213
The past decade has witnessed a great deal of research interest on dynamical processes in complex networks, and most of the focus has been on individual networks that live in isolation. However, many real-world systems consist of a web of interacting networks that are coupled together. Examples include the power grid, the Internet, the transportation network, and the global financial network. Therefore, there is a need for studying the effect of coupling on various network properties, and this has been an active research area in the past two years.
In this presentation, I will review recent results concerning several dynamical processes on ‘coupled’ networks. First, I will consider cascading failures in interdependent networks, where nodes in one network depend on the nodes of the other network and vice versa. I will present results for the critical fraction of nodes whose failure will lead to the collapse of the entire system, in the process revealing the trade-off between network robustness and number of inter-connections (i.e., resources) allocated. Using a toy network model, I will then characterize the optimal inter-link allocation strategy when intra-topologies are unknown. Second, I will discuss the diffusion of information in coupled social-physical networks, and show how the existence of multiple online social networks (e.g., Facebook, Twitter, etc.) affects the speed and scale of information diffusion. The last work I will review studies the diffusion of influence in ‘multiplex’ networks, where links are classified into several types. I will introduce a new social contagion model for multiplex networks and report on the corresponding results that provide insight as to why different content (e.g., a political view, a consumer good, or a rumor) may have completely different spreading characteristics over the same network.
Osman Yağan received the B.S. degree in Electrical and Electronics Engineering from Middle East Technical University, Ankara (Turkey) in 2007, and the Ph.D. degree in Electrical and Computer Engineering from University of Maryland at College Park in August 2011. He was a visiting Postdoctoral Scholar at Arizona State University during Fall 2011. Since December 2011, he has been a Postdoctoral Research Fellow in CyLab at Carnegie Mellon University. His research interests include wireless communications, security, random graphs, social and information networks, and cyber-physical systems.