Neural network algorithms as the cutting edge of self-organization processes

Abstract

With the creation of strong AI, not only humanity, but the entire life evolution enters a new era. Life, as the most complex and most amazing self-organization process, is difficult to accurately describe, but attempts to identify the basic conditions for its development have been made for a long time ago. As it became clear about 70 years ago, the main condition for the life development is the knowledge accumulation in DNA, making it possible to reproduce life forms that have undergone evolutionary selection. Many other important conditions for the development of “living” self-organization were put forward, which were better suited as research objects than knowledge. Although with the beginning of the neural network revolution in machine learning, it became possible to study on models the processes of acquiring knowledge about the surrounding world without human participation, the researchers inertia not to include the cognition processes in self-organization studies remains. A comparison of traditional and neural network approaches to describe self-organization processes was made. Since only the neural nets describe the acquiring knowledge processes, it is concluded that neural network algorithms constitute the cutting edge of the self-organization processes development.