Neural Networks as Surrogate Models for Measurements in Optimization Algorithms

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985807%3A_____%2F10%3A00345993" target="_blank" >RIV/67985807:_____/10:00345993 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Neural Networks as Surrogate Models for Measurements in Optimization Algorithms

Popis výsledku v původním jazyce

The paper deals with surrogate modelling, a modern approach to the optimization of objective functions evaluated via measurements. The approach leads to a substantial decrease of time and costs of evaluation of the objective function, a property that isparticularly attractive in evolutionary optimization. The paper recalls common strategies for using surrogate models in evolutionary optimization, and proposes two extensions to those strategies - extension to boosted surrogate models and extension to using a set of models. These are currently being implemented, in connection with surrogate modelling based on feed-forward neural networks, in a software tool for problem-tailored evolutionary optimization of catalytic materials. The paper presents resultsof experimentally testing already implemented parts and comparing boosted surrogate models with models without boosting, which clearly confirms the usefulness of both proposed extensions.

Název v anglickém jazyce

Neural Networks as Surrogate Models for Measurements in Optimization Algorithms

Popis výsledku anglicky

The paper deals with surrogate modelling, a modern approach to the optimization of objective functions evaluated via measurements. The approach leads to a substantial decrease of time and costs of evaluation of the objective function, a property that isparticularly attractive in evolutionary optimization. The paper recalls common strategies for using surrogate models in evolutionary optimization, and proposes two extensions to those strategies - extension to boosted surrogate models and extension to using a set of models. These are currently being implemented, in connection with surrogate modelling based on feed-forward neural networks, in a software tool for problem-tailored evolutionary optimization of catalytic materials. The paper presents resultsof experimentally testing already implemented parts and comparing boosted surrogate models with models without boosting, which clearly confirms the usefulness of both proposed extensions.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

<a href="/cs/project/GA201%2F08%2F0802" target="_blank" >GA201/08/0802: Aplikace metod znalostního inženýrství při dobývání znalostí z databází</a><br>

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2010

Kód důvěrnosti údajů

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

Název statě ve sborníku

Analytical and Stochastic Modeling Techniques and Applications

ISBN

978-3-642-13567-5

ISSN

—

e-ISSN

—

Počet stran výsledku

16

Strana od-do

—

Název nakladatele

Springer

Místo vydání

Berlin

Místo konání akce

Cardiff

Datum konání akce

14. 6. 2010

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000279619100025