Advanced search
Start date
Betweenand


Lagrangian relaxation and cutting planes for the vertex separator problem

Full text
Author(s):
Cavalcante, Victor F. ; de Souza, Cid C. ; Chen, B ; Paterson, M ; Zhang, G
Total Authors: 5
Document type: Journal article
Source: Lecture Notes in Computer Science; v. 4614, p. 2-pg., 2007-01-01.
Abstract

In this paper we propose an algorithm for the vertex separator problem (VSP) based on Lagrangian Relaxation and on cutting planes derived from valid inequalities presented in [3]. The procedure is used as a preprocessing for the branch-and-cut (B&C) algorithm implemented for VSP in [12], aiming to generate an initial pool of cutting planes and a strong primal bound for the latter. Computational experiments show that the exact method obtained in that way outperforms the pure B&C algorithm recently introduced by de Souza and Balas in [12]. Besides, we show that the Lagrangian phase is a very effective heuristic for the VSP, often producing optimal solutions extremely fast. (AU)

FAPESP's process: 03/09925-5 - Foundations of computer science: combinatory algorithms and discrete structures
Grantee:Yoshiharu Kohayakawa
Support Opportunities: PRONEX Research - Thematic Grants