Diffusion Modelling Topographic Error of SOM Under Control

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21340%2F22%3A00359863" target="_blank" >RIV/68407700:21340/22:00359863 - isvavai.cz</a>

Výsledek na webu

<a href="https://doi.org/10.1007/s11063-021-10660-1" target="_blank" >https://doi.org/10.1007/s11063-021-10660-1</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1007/s11063-021-10660-1" target="_blank" >10.1007/s11063-021-10660-1</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Diffusion Modelling Topographic Error of SOM Under Control

Popis výsledku v původním jazyce

The traditional self-organized map (SOM) is learned by Kohonen learning and the most common 2-dimensional grids defining the structure of the map are the hexagonal grid and the rectangular grid. A novel model of self-organization is based on hexagonal grid and diffusion modeling in continuous space which is a good approximation of endorphins propagation and nitric oxide generation in the real brain. Therefore the structure of the system is described by neuron coordinates instead of neighborhood relationships in traditional SOM. The discussed neuron activation using the diffusion process and novel diffusive learning algorithm is based on this activation mentioned above. The novel structure and algorithm are demonstrated on simple examples and real economic applications.

Název v anglickém jazyce

Diffusion Modelling Topographic Error of SOM Under Control

Popis výsledku anglicky

The traditional self-organized map (SOM) is learned by Kohonen learning and the most common 2-dimensional grids defining the structure of the map are the hexagonal grid and the rectangular grid. A novel model of self-organization is based on hexagonal grid and diffusion modeling in continuous space which is a good approximation of endorphins propagation and nitric oxide generation in the real brain. Therefore the structure of the system is described by neuron coordinates instead of neighborhood relationships in traditional SOM. The discussed neuron activation using the diffusion process and novel diffusive learning algorithm is based on this activation mentioned above. The novel structure and algorithm are demonstrated on simple examples and real economic applications.

Druh

J_imp - Článek v periodiku v databázi Web of Science

CEP obor

—

OECD FORD obor

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Projekt

<a href="/cs/project/EF16_019%2F0000765" target="_blank" >EF16_019/0000765: Výzkumné centrum informatiky</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)<br>S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2022

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název periodika

Neural Processing Letters

ISSN

1370-4621

e-ISSN

1573-773X

Svazek periodika

54

Číslo periodika v rámci svazku

2

Stát vydavatele periodika

CH - Švýcarská konfederace

Počet stran výsledku

18

Strana od-do

835-852

Kód UT WoS článku

000713078000001

EID výsledku v databázi Scopus

2-s2.0-85118305719