Title
A Linear Round Lower Bound for Lovasz-Schrijver SDP Relaxations of Vertex Cover
Abstract
We study semidefinite programming relaxations of Vertex Cover arising from repeated applications of the LS+ "lift-and-project" method of Lovasz and Schrijver starting from the standard linear programming relaxation. Goemans and Kleinberg prove that after one round of LS+ the integrality gap remains arbitrarily close to 2. Charikar proves an integrality gap of 2, later strengthened by Hatami, Magen, and Markakis, for stronger relaxations that are, however, incomparable with two rounds of LS+. Subsequent work by Georgiou, Magen, Pitassi, and Tourlakis shows that the integrality gap remains 2 - \varepsilon after \Omega(\sqrt {\frac{{\log n}} {{\log \log n}}} ) rounds [?]. We prove that the integrality gap remains at least 7/6 - \varepsilon after c_{\varepsilon}n rounds, where n is the number of vertices and c_\varepsilon \ge 0 is a constant that depends only on \varepsilon.
Year
DOI
Venue
2007
10.1109/CCC.2007.2
CCC '07 Proceedings of the Twenty-Second Annual IEEE Conference on Computational Complexity
Keywords
Field
DocType
linear programming,lovasz-schrijver sdp relaxations,integrality gap,lift-and-project method,linear round lower bound,semidefinite programming relaxations,standard linear programming relaxation,vertex cover,linear programming relaxation,lower bound,projection method
Discrete mathematics,Combinatorics,Upper and lower bounds,Vertex cover,Linear programming relaxation,Mathematics
Conference
ISSN
ISBN
Citations 
1093-0159
0-7695-2780-9
17
PageRank 
References 
Authors
1.20
2
3
Name
Order
Citations
PageRank
Grant Schoenebeck150939.48
Luca Trevisan22995232.34
Madhur Tulsiani335824.60