The monograph examines: the history of artificial neural networks (INS), classification of INS (according to modularity, connectivity, training, tasks to be solved, presence of hidden layers, time delays and feedback) with examples, ANN tuning technology (algorithms for selecting informative factors and reducing the dimension of data, methods for normalizing informative factors, functions for activating artificial neurons and their features, estimates of the power of the training set, rules for training ANN weights, conditions for completing training), static INS without associative memory are given (MLP, RBFNN, GRNN, PNN, SVM, etc.), static INS with associative memory (SOM, LVQNN, CPNN, etc.), non-recurrent dynamic INS (NARNN, NARMANN, TDNN, etc.), recurrent INS without associative memory (ENN, RMLP, etc.), recurrent ANNs with associative memory (HNN, etc.) and their comparative characteristics