Train the Multi-Layer Perceptrons Based on Crow Search Algorithm

This paper presents a new approach for training the Multi-Layer Perceptron (MLP) based on the crow search algorithm (CSA) optimization algorithm. The main objective of this approach is to reduce the error to its minimum level and increase the rate of the classification. The benchmark of the proposed approach implementation achieved using more than one standard datasets for classification to ensure the quality of the result, and the performance also compared with other optimization algorithms such as Particle Swarm optimization (PSO), Genetic Algorithm (GA), and Ant Colony optimization (ACO). The outcomes indicated that the crow search algorithm was the best since it produces the highest accuracy rate and solves the optimization problem efficiently.