ON MATHEMATICAL MODELS OF ARTIFICIAL NEURAL NETWORKS

Inna Samuilik; Felix Sadyrbaev; Svetlana Atslēga

doi:10.22616/ERDev.2023.22.TF007

ON MATHEMATICAL MODELS OF ARTIFICIAL NEURAL NETWORKS

Inna Samuilik
, Felix Sadyrbaev
, Svetlana Atslēga

University of Latvia

Riga Technical University
University of Latvia
Latvia University of Life Sciences and Technologies

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › Research › peer-review

5 Citations (Scopus)

Abstract

Artificial Neural Networks are in focus. The four-dimensional and the five-dimensional systems are considered. The activation function – hyperbolic tangent is used to model the Artificial Neural Networks. By changing one of the parameters in the system, different types of solutions are obtained: periodic solutions and chaotic solutions. The graphs of all solutions, the dynamics of Lyapunov exponents and 2D and 3D projections of attractors are provided using Wolfram Mathematica.

Original language	English
Title of host publication	Engineering for Rural Development
Place of Publication	Jelgava
Publisher	Latvia University of Life Sciences and Technologies
Pages	45-50
Number of pages	6
Volume	22
DOIs	https://doi.org/10.22616/ERDev.2023.22.TF007
Publication status	Published - 2023

Publication series

Name	Engineering for Rural Development
Publisher	Latvia University of Life Sciences and Technologies
ISSN (Print)	1691-3043

Keywords

Artificial Neural Networks
chaotic solutions
Lyapunov exponents
periodic solutions

OECD Field of Science

1.1 Mathematics

Access to Document

10.22616/ERDev.2023.22.TF007

Cite this

@inproceedings{2efd37b38878437391af8974b2721b49,

title = "ON MATHEMATICAL MODELS OF ARTIFICIAL NEURAL NETWORKS",

abstract = "Artificial Neural Networks are in focus. The four-dimensional and the five-dimensional systems are considered. The activation function – hyperbolic tangent is used to model the Artificial Neural Networks. By changing one of the parameters in the system, different types of solutions are obtained: periodic solutions and chaotic solutions. The graphs of all solutions, the dynamics of Lyapunov exponents and 2D and 3D projections of attractors are provided using Wolfram Mathematica.",

keywords = "Artificial Neural Networks, chaotic solutions, Lyapunov exponents, periodic solutions",

author = "Inna Samuilik and Felix Sadyrbaev and Svetlana Atslēga",

year = "2023",

doi = "10.22616/ERDev.2023.22.TF007",

language = "English",

volume = "22",

series = "Engineering for Rural Development",

publisher = "Latvia University of Life Sciences and Technologies",

pages = "45--50",

booktitle = "Engineering for Rural Development",

address = "Latvia",

}

ON MATHEMATICAL MODELS OF ARTIFICIAL NEURAL NETWORKS. / Samuilik, Inna; Sadyrbaev, Felix; Atslēga, Svetlana.
Engineering for Rural Development. Vol. 22 Jelgava: Latvia University of Life Sciences and Technologies, 2023. p. 45-50 (Engineering for Rural Development).

Research output: Chapter in Book/Report/Conference proceeding › Conference paper › Research › peer-review

TY - GEN

T1 - ON MATHEMATICAL MODELS OF ARTIFICIAL NEURAL NETWORKS

AU - Samuilik, Inna

AU - Sadyrbaev, Felix

AU - Atslēga, Svetlana

PY - 2023

Y1 - 2023

N2 - Artificial Neural Networks are in focus. The four-dimensional and the five-dimensional systems are considered. The activation function – hyperbolic tangent is used to model the Artificial Neural Networks. By changing one of the parameters in the system, different types of solutions are obtained: periodic solutions and chaotic solutions. The graphs of all solutions, the dynamics of Lyapunov exponents and 2D and 3D projections of attractors are provided using Wolfram Mathematica.

AB - Artificial Neural Networks are in focus. The four-dimensional and the five-dimensional systems are considered. The activation function – hyperbolic tangent is used to model the Artificial Neural Networks. By changing one of the parameters in the system, different types of solutions are obtained: periodic solutions and chaotic solutions. The graphs of all solutions, the dynamics of Lyapunov exponents and 2D and 3D projections of attractors are provided using Wolfram Mathematica.

KW - Artificial Neural Networks

KW - chaotic solutions

KW - Lyapunov exponents

KW - periodic solutions

UR - https://www.tf.lbtu.lv/conference/proceedings2023/Papers/TF007.pdf

UR - https://www.scopus.com/pages/publications/85169903227

U2 - 10.22616/ERDev.2023.22.TF007

DO - 10.22616/ERDev.2023.22.TF007

M3 - Conference paper

VL - 22

T3 - Engineering for Rural Development

SP - 45

EP - 50

BT - Engineering for Rural Development

PB - Latvia University of Life Sciences and Technologies

CY - Jelgava

ER -

ON MATHEMATICAL MODELS OF ARTIFICIAL NEURAL NETWORKS

Abstract

Publication series

Keywords

OECD Field of Science

Access to Document

Other files and links

Fingerprint

Cite this