Počet záznamů: 1
Evolutionary Learning of Regularization Networks with Multi-kernel Units
- 1.
SYSNO ASEP 0369159 Druh ASEP C - Konferenční příspěvek (mezinárodní konf.) Zařazení RIV D - Článek ve sborníku Název Evolutionary Learning of Regularization Networks with Multi-kernel Units Tvůrce(i) Vidnerová, Petra (UIVT-O) RID, SAI, ORCID
Neruda, Roman (UIVT-O) SAI, RID, ORCIDZdroj.dok. Advances in Neural Networks – ISNN 2011. Part I. - Berlin : Springer, 2011 / Liu D. ; Zhang H. ; Polycarpou M. ; Alippi C. ; He H. - ISSN 0302-9743 - ISBN 978-3-642-21104-1 Rozsah stran s. 538-546 Poč.str. 9 s. Akce ISNN 2011. International Symposium on Neural Networks /8./ Datum konání 29.05.2011-01.06.2011 Místo konání Guilin Země CN - Čína Typ akce WRD Jazyk dok. eng - angličtina Země vyd. DE - Německo Klíč. slova genetic algorithms ; kernel functions ; regularization networks Vědní obor RIV IN - Informatika CEP GAP202/11/1368 GA ČR - Grantová agentura ČR CEZ AV0Z10300504 - UIVT-O (2005-2011) UT WOS 000301802600062 EID SCOPUS 79957795865 DOI 10.1007/978-3-642-21105-8_62 Anotace Regularization networks represent an important supervised learning method applicable for regression and classification tasks. They benefit from very good theoretical background, although the presence of meta parameters is their drawback. The meta parameters, including the type of kernel function, are typically supposed to be given in advance and come ready as an input of the algorithm. In this paper, we propose multi-kernel functions, namely product kernel functions and composite kernel functions. The choice of kernel function becomes part of the optimization process, for which a new evolutionary learning algorithm is introduced that deals with different kernel functions, including composite kernels. The results are demonstrated on experiments with benchmark tasks. Pracoviště Ústav informatiky Kontakt Tereza Šírová, sirova@cs.cas.cz, Tel.: 266 053 800 Rok sběru 2012
Počet záznamů: 1