NEAT in HyperNEAT Substituted with Genetic Programming

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F09%3A00159310" target="_blank" >RIV/68407700:21230/09:00159310 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

NEAT in HyperNEAT Substituted with Genetic Programming

Popis výsledku v původním jazyce

In this paper we present application of genetic programming (GP) to evolution of indirect encoding of neural network weights. We compare usage of original HyperNEAT algorithm with our implementation, in which we replaced the underlying NEAT with geneticprogramming. The algorithm was named HyperGP. The evolved neural networks were used as controllers of autonomous mobile agents (robots) in simulation. The agents were trained to drive with maximum average speed. This forces them to learn how to drive onroads and avoid collisions. The genetic programming lacking the NEAT complexification property shows better exploration ability and tends to generate more complex solutions in fewer generations. On the other hand, the basic genetic programming generatesquite complex functions for weights generation. Both approaches generate neural controllers with similar abilities.

Název v anglickém jazyce

NEAT in HyperNEAT Substituted with Genetic Programming

Popis výsledku anglicky

In this paper we present application of genetic programming (GP) to evolution of indirect encoding of neural network weights. We compare usage of original HyperNEAT algorithm with our implementation, in which we replaced the underlying NEAT with geneticprogramming. The algorithm was named HyperGP. The evolved neural networks were used as controllers of autonomous mobile agents (robots) in simulation. The agents were trained to drive with maximum average speed. This forces them to learn how to drive onroads and avoid collisions. The genetic programming lacking the NEAT complexification property shows better exploration ability and tends to generate more complex solutions in fewer generations. On the other hand, the basic genetic programming generatesquite complex functions for weights generation. Both approaches generate neural controllers with similar abilities.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2009

Kód důvěrnosti údajů

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

Název statě ve sborníku

Adaptive and Natural Computing Algorithms

ISBN

978-3-642-04920-0

ISSN

0302-9743

e-ISSN

—

Počet stran výsledku

10

Strana od-do

—

Název nakladatele

Springer

Místo vydání

Heidelberg

Místo konání akce

Kuopio

Datum konání akce

23. 4. 2009

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—