Publication detail

Models of Iterated Artificial Neurons

SMETANA, B. CHVALINA, J.

Original Title

Models of Iterated Artificial Neurons

English Title

Models of Iterated Artificial Neurons

Type

conference paper

Language

en

Original Abstract

Abstract. Application of algebraic structures in research of structure of the most used artificial neural network - multilayer perceptron and functionality of artificial neuron, is one from current the most interesting areas in the usage of time varying artificial neurons. In this paper, we have described certain properties of constructed algebraic structures of artificial neurons, including the hypergroup formed by iterated models of artificial neurons. We describe constructions of P-hypergroups - a certain special case of which are variants of semigroups - of artificial time-varying neurons based on investigations of Vougiouklis - Konguetsof and also construction of the cascade with the phase set of neurons. These concepts yield a base for the building of algebraic systems of artificial neurons which deserves to be developed.

English abstract

Abstract. Application of algebraic structures in research of structure of the most used artificial neural network - multilayer perceptron and functionality of artificial neuron, is one from current the most interesting areas in the usage of time varying artificial neurons. In this paper, we have described certain properties of constructed algebraic structures of artificial neurons, including the hypergroup formed by iterated models of artificial neurons. We describe constructions of P-hypergroups - a certain special case of which are variants of semigroups - of artificial time-varying neurons based on investigations of Vougiouklis - Konguetsof and also construction of the cascade with the phase set of neurons. These concepts yield a base for the building of algebraic systems of artificial neurons which deserves to be developed.

Keywords

P-hypergroups, iterated models, artificial neurons, groups of neurons.

Released

05.02.2019

Publisher

Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering

Location

Bratislava

ISBN

978-80-227-4884-1

Book

18th CONFERENCE ON APPLIED MATHEMATICS APLIMAT 2019 PROCEEDINGS

Pages from

203

Pages to

2012

Pages count

10

URL

Documents

BibTex


@inproceedings{BUT156491,
  author="Jan {Chvalina} and Bedřich {Smetana}",
  title="Models of Iterated Artificial Neurons",
  annote="Abstract. Application of algebraic structures in research of structure of the most used artificial neural network - multilayer perceptron and functionality of artificial neuron, is one from current the most interesting areas in the usage of time varying artificial neurons. In this paper, we have described certain properties of constructed algebraic structures of artificial neurons, including the hypergroup formed by iterated models of artificial neurons. We describe constructions of P-hypergroups - a certain special case of which are variants of semigroups - of artificial time-varying neurons based on investigations of Vougiouklis - Konguetsof and also construction of the cascade with the phase set of neurons. These concepts yield a base for the building of algebraic systems of artificial neurons which deserves to be developed.",
  address="Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering",
  booktitle="18th CONFERENCE ON APPLIED MATHEMATICS APLIMAT 2019 PROCEEDINGS",
  chapter="156491",
  howpublished="electronic, physical medium",
  institution="Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering",
  year="2019",
  month="february",
  pages="203--2012",
  publisher="Slovak University of Technology in Bratislava, Faculty of Mechanical Engineering",
  type="conference paper"
}