Multi-objectivization and Surrogate Modelling for Neural Network Hyper-parameters Tuning
The result's identifiers
Result code in IS VaVaI
<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F13%3A10173118" target="_blank" >RIV/00216208:11320/13:10173118 - isvavai.cz</a>
Alternative codes found
RIV/67985807:_____/13:00428789
Result on the web
<a href="http://link.springer.com/chapter/10.1007/978-3-642-39678-6_11" target="_blank" >http://link.springer.com/chapter/10.1007/978-3-642-39678-6_11</a>
DOI - Digital Object Identifier
<a href="http://dx.doi.org/10.1007/978-3-642-39678-6_11" target="_blank" >10.1007/978-3-642-39678-6_11</a>
Alternative languages
Result language
angličtina
Original language name
Multi-objectivization and Surrogate Modelling for Neural Network Hyper-parameters Tuning
Original language description
We present a multi-objectivization approach to the parameter tuning of RBF networks and multilayer perceptrons. The approach works by adding two new objectives - maximization of kappa statistic and minimization of root mean square error - to the originally single-objective problem of minimizing the classification error of the model. We show the performance of the multiobjectivization approach on five data sets and compare it to a surrogate based single-objective algorithm for the same problem. Moreover,we compare the multi-objectivization approach to two surrogate based approaches - a singleobjective one and a multi-objective one.
Czech name
—
Czech description
—
Classification
Type
D - Article in proceedings
CEP classification
IN - Informatics
OECD FORD branch
—
Result continuities
Project
<a href="/en/project/LD13002" target="_blank" >LD13002: Modeling of complex systems for softcomputing methods</a><br>
Continuities
S - Specificky vyzkum na vysokych skolach
Others
Publication year
2013
Confidentiality
S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů
Data specific for result type
Article name in the collection
Communications in Computer and Information Science
ISBN
978-3-642-39677-9
ISSN
1865-0929
e-ISSN
—
Number of pages
7
Pages from-to
61-66
Publisher name
Springer Berlin Heidelberg
Place of publication
Berlin, Německo
Event location
Nanning, China
Event date
Jul 28, 2013
Type of event by nationality
WRD - Celosvětová akce
UT code for WoS article
—