Most machine learning problems deal with datasets in which every of the given examples consist of a large list of features. Usually, the set of features comes without knowledge about the discriminative information provided by each element. In this scenario, feature selection is essential to perform the classification task with reduced complexity and acceptable performance. A key issue is to define a criterion in order to rank the features, discarding those features that are less relevant, redundant, or noisy. This depends on the particular task, the classifier and the properties of the data. Another key issue comes from the infeasibility of evaluating all possible combinations, which is why an intelligent search strategy is required for finding an optimal subset. The objective of feature selection is the improvement of a machine learning model, either in terms of learning speed, computational complexity, simplicity, interpretability of the representation, and/or generalization capability.