Use of HyperNEAT Encoding for Hybrid Arti?cial Neural Networks

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F15%3A00236679" target="_blank" >RIV/68407700:21230/15:00236679 - isvavai.cz</a>

Výsledek na webu

<a href="https://www.researchgate.net/publication/285720854_Use_of_HyperNEAT_Encoding_for_Hybrid_Artificial_Neural_Networks" target="_blank" >https://www.researchgate.net/publication/285720854_Use_of_HyperNEAT_Encoding_for_Hybrid_Artificial_Neural_Networks</a>

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Use of HyperNEAT Encoding for Hybrid Arti?cial Neural Networks

Popis výsledku v původním jazyce

In the recent years there has been significant upturn in the development of hybrid modular systems, with the emphasis on solving various problems in the Artificial Intelli- gence domain. These hybrid neural network systems outperform artificial neural networks in distinct areas of the computer science research. In order for these systems to function properly, it is important to interconnect the system?s nodes in valid, or in the best case optimal topology. Large amount of algorithms for optimization ofthe artificial neural networks exist, but not many operate, or have been modified to operate on these hybrid modular systems. The analytical solutions, are computationaly complex and therefore scientists design and implement evolutionary optimization algorithms to achieve satis- factory results. The novel algorithm, representing the state of the art of the indirect encoding-based neuro-evolutionary algorithms is HyperNEAT algorithm. The aim of this thesis is to design and implement exten

Název v anglickém jazyce

Use of HyperNEAT Encoding for Hybrid Arti?cial Neural Networks

Popis výsledku anglicky

In the recent years there has been significant upturn in the development of hybrid modular systems, with the emphasis on solving various problems in the Artificial Intelli- gence domain. These hybrid neural network systems outperform artificial neural networks in distinct areas of the computer science research. In order for these systems to function properly, it is important to interconnect the system?s nodes in valid, or in the best case optimal topology. Large amount of algorithms for optimization ofthe artificial neural networks exist, but not many operate, or have been modified to operate on these hybrid modular systems. The analytical solutions, are computationaly complex and therefore scientists design and implement evolutionary optimization algorithms to achieve satis- factory results. The novel algorithm, representing the state of the art of the indirect encoding-based neuro-evolutionary algorithms is HyperNEAT algorithm. The aim of this thesis is to design and implement exten

Druh

O - Ostatní výsledky

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2015

Kód důvěrnosti údajů

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