CONSTRUCTION OF NEURAL NETWORKS FOR CLASSIFICATION POINTS OF LINEAR COMPLEX OF PLANES OF GENERAL TYPE

Abstract

The work describes the construction, training and testing of an artificial neural network (ANN) designed for determining whether an arbitrary point a seven-dimensional projective complex spacebelongs to the variety of singular points of the first kind for linear complex of planes of general type. This complex does not admit the singular points of the second kind and the completely singular planes.

Keywords

• тривектор
• арифметическая характеристика тривектора
• линейный комплекс плоскостей
• многообразие особых точек
• нейронная сеть прямого распространения
• trivector
• arifmetic characteristic of trivector
• linear complex of planes
• the variety of singular points
• the feedforward neural network