Dynamical systems approach to outlier robust deep neural networks for regression for
regression based on the gradient conjugate prior (GCP) updates. We show that contaminating the training data
A method for separation of classes of linear conjugation problems for three-dimensional vector problems of fractional-
linear conjugation. The main result of the paper is the following one: known two
Method of Binary Analytic Programming to Look for Optimal Mathematical ExpressionIn the known methods of symbolical
regression by search of the solution with the help of a genetic