An approach for the knapsack problem using genetic algorithms with learning capabilities

In this paper we used a genetic algorithm with learning capabilities to approach a problem that is NP-hard by definition known as the knapsack problem. The experiments made showed that a genetic algorithm that is able to learn reaches a solution faster than a simple genetic algorithm. The drawback of any heuristic approach like this is that the result obtained often is a local maximum. The learning process can be a good aid when we need to reach a solution rapidly, but we have to keep in mind that this solution may not be the best one available.