Better Balance By Being Biased: Approximation Algorithms For Max Bisection

Lecturer : 
Per Austrin
Event type: 
HIIT seminar
Doctoral dissertation
Respondent: 
Opponent: 
Custos: 
Event time: 
2012-10-15 13:15 to 14:00
Place: 
Lecture Hall T2, Konemiehentie 2
Description: 

Abstract:
I'll discuss approximation algorithms for the Max Bisection problem, culminating in a recent result (joint with Siavosh Benabbas and Konstantinos Georgiou, to appear in SODA 2013) giving a near-optimal approximation ratio.

Bio:
Per Austrin received his PhD in 2008 from the Royal Institute of Technology in Stockholm, Sweden. Following this he was a post-doc at Institut Mittag-Leffler, New York University, and University of Toronto, and is now an Aalto Science Fellow.  His research area is computational complexity, with a focus on approximation algorithms and their limitations.

Host: Pekka Orponen


Last updated on 9 Oct 2012 by Sohan Seth - Page created on 9 Oct 2012 by Sohan Seth