Solving k-cluster problems to optimality using adjustable semidefinite programming bounds - Archive ouverte HAL Accéder directement au contenu
Rapport (Rapport De Recherche) Année : 2010

Solving k-cluster problems to optimality using adjustable semidefinite programming bounds

Jérôme Malick

Résumé

This paper deals with computing exact solutions of a classical NP-hard problem of combinatorial optimization, the k-cluster problem. This problem consists in finding a heaviest subgraph with k nodes in a weighted graph, (or a densest subgraph, if the graph is unweighted). We present a branch-and-bound approach that applies a novel bounding procedure, based on recent semidefinite programming techniques. We use new semidefinite bounds that are less accurate than the standard semidefinite bounds, but cheaper to get. The experiments show that this approach is competitive with the best existing ones.
Fichier non déposé

Dates et versions

hal-01125758 , version 1 (06-03-2015)

Identifiants

  • HAL Id : hal-01125758 , version 1

Citer

Jérôme Malick, Frédéric Roupin. Solving k-cluster problems to optimality using adjustable semidefinite programming bounds. [Research Report] CEDRIC-10-1920, CEDRIC Lab/CNAM. 2010. ⟨hal-01125758⟩
31 Consultations
0 Téléchargements

Partager

Gmail Facebook X LinkedIn More