An elitist genetic algorithm for the maximum independent set problem

Abstract

Genetic algorithms are a computational paradigm belonging to the class of optimization techniques known as evolutionary computation. They have been implemented successfully to solve many difficult optimization problems. We have developed a new genetic algorithm for the maximum independent set problem based on the elitist strategy. The algorithm presented is tested on the so-called DIMACS benchmark graphs. The effectiveness of the algorithm is very satisfactory since it outperforms in most cases the genetic algorithms for the maximum independent set problem reported in the literature.