Performance and Scaling in Complex Networks: A Spectral Approach
Milena Mihail College of Computing Georgia Institute of Technology
Monday, May 23, 2005 JEC 3117 - 4:00 p.m. to 5:00 p.m. Refreshments at 3:30 p.m.
The issue of scalability is of fundamental importance in complex networks. In the context of communication networks the questions concern network performance. How does congestion scale on the Internet? At what rate do crawlers discover new web pages? In the context of biological networks the questions concern reverse engineering. What evolutionary primitives have driven the phenotype of a particular gene or protein interaction network? What are the most significant nodes? Spectral graph theory is a powerful mathematical tool that characterizes the performance of many algorithms run on a particular graph. We use spectral graph theory to answer some of the above questions.
Bio: Milena Mihail is Associate Professor of Computer Science
Tech since 1999. Her research areas are Theory and Networking. Prior to
joining Georgia Tech, she worked
for 10 years at Bellcore, initially as Member of Technical
Staff and subsequently as Director. She has also held
visiting positions at Columbia University and U.C. Berkeley.
She received her Ph.D. from Harvard University in 1989,
and her Diploma in Electrical Engineering from the
Polytechnic of Athens, Greece in 1984.
Last updated: May 16, 2005