Modeling of stochastic brain function in artificial intelligence

Abstract

Objectives -research of stochastic brain function in respect to creation of artificial intelligence. Material and methods. Mathematical modeling principles were used for simulation of brain functioning in a stochastic mode. Results. Two types of brain activity were considered: determinated type, usually modeled using the perceptron, and stochastic type. It is shown, that stochastic brain function modeling is the necessary condition for AI to become capable of creativity, generation of new knowledge. Mathematical modeling of a neural network of the cerebral cortex, consisting of the set of the cyclic neuronal circuits (memory units), was performed for the stochastic mode of brain functioning. Models of "two-dimensional" and "one-dimensional" brain were analyzed. The pattern of excitation in memory units was calculated in the "one-dimensional" brain model. Conclusion. Relying on the knowledge of the stochastic mode of brain function, a way of creation of AI can be offered. а-rhythm of a patient is a recommended focus of the therapist's attention in diagnostics and treatment of brain disorders. It was noted, that the alpha wave amplitude and frequency could indicate the cognitive, creative and intuitive abilities of a person.