Intensification/Diversification-Driven ILS for a Graph Coloring Problem

Samir Loudni (1)

1 Equipe CODAG - Laboratoire GREYC - UMR6072

This paper presents an extension of the ILS algorithm, called ID-ILS, by introducing new local search devices that enforce an efficient tradeoff of intensification and diversification. Experiments performed on the DIMACS benchmarks show that our method is competitive with the best coloring algorithms.

Domaines: Intelligence artificielle [cs.AI]; Traitement du texte et du document

Samir Loudni. Intensification/Diversification-Driven ILS for a Graph Coloring Problem. 12th European Conference on Evolutionary Computation in Combinatorial Optimization (EvoCOP 2012), Apr 2012, Malaga, Spain. pp.160-171. ⟨hal-01022859⟩