George Kesidis, EE and CS&E Departments, Pennsylvania State Univer

Location :

Thomson

46 Quai A. Le Gallo
92648 Boulogne
France

Building: Campus, 
Room: Pantheon,

Abstract :

We begin with a brief overview of incentive systems to encourage
cooperative behavior in peer-to-peer networking activity.
Reputation systems are often used to quantify observations of peer
behavior, e.g., a simple reputation system for a file-sharing game.
Referral systems can be used when direct observations of peer
activity is lacking.  Reputation-referral systems can, however,
be manipulated for selfish or malicious purposes.  We explain how
a chain-referral system that multiplies the component ``direct"
reputations is immune to sybil attacks. The talk concludes with
a brief discussion of non-sybil false referrals and implementation
issues.

Speaker's bio:

George Kesidis received his M.S. and Ph.D. in EECS 
from U.C. Berkeley in 1990 and 1992 respectively. 
He was a professor in the E&CE Dept of the University
of Waterloo, Canada, from 1992 to 2000. Since April 2000,
he has been a professor in both the EE and CS&E 
Depts of the Pennsylvania State University. His research
includes several areas of computer/communication
networking and, more recently, machine learning.
He served as co-chair of IEEE INFOCOM 2007
and is currently on the editorial board of ACM TOMACS
and IEEE Communications Surveys and Tutorials. 
He is a senior member of the IEEE.

Host :

Theodoros Salonidis, Thomson