Meta-learning approach to neural network optimization

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21230%2F10%3A00167763" target="_blank" >RIV/68407700:21230/10:00167763 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/68407700:21240/10:00167763

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Meta-learning approach to neural network optimization

Popis výsledku v původním jazyce

Optimization of neural network topology, weights and neuron transfer functions for given data set and problem is not an easy task. In this article, we focus primarily on building optimal feed-forward neural network classifier for i.i.d. data sets. We apply metalearning principles to the neural network structure and function optimization. We show that diversity promotion, ensembling, self-organization and induction are beneficial for the problem. We combine several different neuron types trained by various optimization algorithms to build a supervised feedforward neural network called Group of Adaptive Models Evolution (GAME). The approach was tested on wide number of benchmark data sets. The experiments show that the combination of different optimization algorithms in the network is the best choice when the performance is averaged over several real-world problems.

Název v anglickém jazyce

Meta-learning approach to neural network optimization

Popis výsledku anglicky

Optimization of neural network topology, weights and neuron transfer functions for given data set and problem is not an easy task. In this article, we focus primarily on building optimal feed-forward neural network classifier for i.i.d. data sets. We apply metalearning principles to the neural network structure and function optimization. We show that diversity promotion, ensembling, self-organization and induction are beneficial for the problem. We combine several different neuron types trained by various optimization algorithms to build a supervised feedforward neural network called Group of Adaptive Models Evolution (GAME). The approach was tested on wide number of benchmark data sets. The experiments show that the combination of different optimization algorithms in the network is the best choice when the performance is averaged over several real-world problems.

Druh

J_x - Nezařazeno - Článek v odborném periodiku (Jimp, Jsc a Jost)

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

<a href="/cs/project/KJB201210701" target="_blank" >KJB201210701: Automatická extrakce znalostí</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 periodika

Neural Networks

ISSN

0893-6080

e-ISSN

—

Svazek periodika

2010 (23)

Číslo periodika v rámci svazku

4

Stát vydavatele periodika

NL - Nizozemsko

Počet stran výsledku

15

Strana od-do

—

Kód UT WoS článku

000277227900013

EID výsledku v databázi Scopus

—