… Particle Swarm PSO and reduced PSO algorithm. The first initialization method is chaotic
initialization with … for PSO algorithm, the second initialization method called reduced size PSO (…

… v 2 i init is the initial velocity of the ith particle of the PSO2 algorithm; … Specifically, the
particle swarm optimization algorithm was used to train the artificial neural networks (PSO-ANN …

… of the proposed approach ANN-PSO, in the two phases of ANN (training and testing)
optimized by PSO algorithm approach, for the most robust and efficient approach model structure, …

… The parameters used in the proposed psoCNN can be grouped in three categories: particle
swarm optimization, CNN architecture initialization, and CNN training. The parameters used …

… This proves the stability of our model, as its performance is due to the architecture and
training and not random causes like the initialization of the learnable parameters. …

… to train neural networks, are prone to local minima entrapment, and for this reason, a strategy
to initialize the training … This algorithm is an evolution of the well-known particle swarm …

… Thus, we propose a novel initialization method for ISBPSO using MI. … is randomly divided
into a training set (70%) and a test set (30%). The training set is fed to a FS method to obtain a …

… and initialize fine… method and traditional method to perform modulation recognition. Although
time complexity is high in the case of large training datasets or particle quantity, the method …

… , a new approach to initialize population is proposed using … at random locations for swarm
initialization. The proposed … [16], for the training of neural networks and to solve non-linear …

… mixed initialization as in [30]. In this initialization method, most particles are initialized with a
… its application in training of feedforward neural networks,’’ Knowl.-Based Syst., vol. 143, pp. …