Building Predictive Models in Two Stages with Meta-Learning Templates optimized by Genetic Programming

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21240%2F14%3A00224815" target="_blank" >RIV/68407700:21240/14:00224815 - isvavai.cz</a>

Výsledek na webu

<a href="http://dx.doi.org/10.1109/CIEL.2014.7015740" target="_blank" >http://dx.doi.org/10.1109/CIEL.2014.7015740</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/CIEL.2014.7015740" target="_blank" >10.1109/CIEL.2014.7015740</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Building Predictive Models in Two Stages with Meta-Learning Templates optimized by Genetic Programming

Popis výsledku v původním jazyce

The model selection stage is one of the most difficult in predictive modeling. To select a model with a highest generalization performance involves benchmarking huge number of candidate models or algorithms. Often, a final model is selected without considering potentially high quality candidates just because there are too many possibilities. Improper benchmarking methodology often leads to biased estimates of model generalization performance. Automation of the model selection stage is possible, howeverthe computational complexity is huge especially when ensembles of models and optimization of input features should be also considered. In this paper we show, how to automate model selection process in a way that allows to search for complex hierarchies of ensemble models while maintaining computational tractability. We introduce two-stage learning, meta-learning templates optimized by evolutionary programming with anytime properties to be able to deliver and maintain data-tailored algori

Název v anglickém jazyce

Building Predictive Models in Two Stages with Meta-Learning Templates optimized by Genetic Programming

Popis výsledku anglicky

The model selection stage is one of the most difficult in predictive modeling. To select a model with a highest generalization performance involves benchmarking huge number of candidate models or algorithms. Often, a final model is selected without considering potentially high quality candidates just because there are too many possibilities. Improper benchmarking methodology often leads to biased estimates of model generalization performance. Automation of the model selection stage is possible, howeverthe computational complexity is huge especially when ensembles of models and optimization of input features should be also considered. In this paper we show, how to automate model selection process in a way that allows to search for complex hierarchies of ensemble models while maintaining computational tractability. We introduce two-stage learning, meta-learning templates optimized by evolutionary programming with anytime properties to be able to deliver and maintain data-tailored algori

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2014

Kód důvěrnosti údajů

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

Název statě ve sborníku

2014 IEEE Symposium on Computational Intelligence in Ensemble Learning (CIEL) Proceedings

ISBN

978-1-4799-4513-9

ISSN

—

e-ISSN

—

Počet stran výsledku

8

Strana od-do

27-34

Název nakladatele

IEEE

Místo vydání

Piscataway

Místo konání akce

Orlando

Datum konání akce

9. 12. 2014

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—