Skip navigation
Please use this identifier to cite or link to this item:
Title: Neural network structures: current and future states
Authors: Davydov, M. V.
Kilin, S. Ya.
Kulchitsky, V. A.
Keywords: материалы конференций;neural network structures
Issue Date: 2018
Publisher: БГУИР
Citation: Davydov, M. V. Neural network structures: current and future states / M. V. Davydov, S. Y. Kilin, V. A. Kulchitsky // Открытые семантические технологии проектирования интеллектуальных систем = Open Semantic Technologies for Intelligent Systems (OSTIS-2018) : материалы международной научно-технической конференции (Минск, 15 - 17 февраля 2018 года) / редкол. : В. В. Голенков (отв. ред.) [и др.]. – Минск : БГУИР, 2018. – С. 259 - 264.
Abstract: Two prevailing points of view on the way of creating artificial intelligence have been presented: based on the consistent patterns the evolution of neurons of the natural nervous system went through and based on radically new patterns, different from the consistent patterns seen in the natural evolution, shall be used to develop neural network structures and programs. The main structural elements of the future artificial intelligence based on the laws of the development of human neural networks have been presented: the main types of neurotransmitters and their role in the modulation of informative signals; a principle of diffuse transmission of information through the intercellular matrix; a concept of improving efficiency of functioning neural network structures based on diffuse transmission of signals in the form of hormones, growth factors, cytokines. It was pointed out that the researchers who are aiming at the development of artificial intelligence must remember about potential negative consequences that should be taken into account and avoided at the stages of problem semantic comprehension.
Appears in Collections:OSTIS-2018

Files in This Item:
File Description SizeFormat 
Davydov_Neural.PDF258.89 kBAdobe PDFView/Open
Show full item record

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.