Application of Negative Learning Ant Colony Optimization to the Far from Most String Problem

Abstract

We propose the application of a recently introduced version of ant colony optimization—negative learning ant colony optimization—to the far from most string problem. This problem is a notoriously difficult combinatorial optimization problem from the group of string selection problems. The proposed algorithm makes use of negative learning in addition to the standard positive learning mechanism in order to achieve better guidance for the exploration of the search space. In addition, we compare different versions of our algorithm characterized by the use of different objective functions. The obtained results show that our algorithm is especially successful for instances with specific characteristics. Moreover, it becomes clear that none of the existing state-of-the-art methods is best for all problem instances.