2.4.1. NSGA-II Algorithm

Non-dominated Sorting Genetic Algorithm-II (NSGA-II) [48,49] has been implemented to deal with the multi-objective optimization problem that is considered in our hyperparameter searching problem. Genetic operators and the codification of the individuals are specific for the problem at hand. This work involves a supervised classification problem, where the individual and its chromosomes correspond to the architecture of a convolutional neural network, which is in charge of the signal classification. The performance evaluation of each individual is determined by the complexity of the CNN and its classification accuracy. Algorithm 1 shows the pseudo-code of the NSGA-II.