ϵ-sparse representations generalized sparse approximation and the equivalent family of SVM tasks /

Relation between a family of generalized Support Vector Machine (SVM) problems and the novel ϵ-sparse representation is provided. In defining ϵ-sparse representations, we use a natural generalization of the classical ϵ-insensitive cost function for vectors. The insensitive parameter of the SVM probl...

Teljes leírás

Elmentve itt :
Bibliográfiai részletek
Szerzők: Szabó Zoltán
Lőrincz András
Dokumentumtípus: Cikk
Megjelent: 2006
Sorozat:Acta cybernetica 17 No. 3
Kulcsszavak:Számítástechnika, Kibernetika
Tárgyszavak:
Online Access:http://acta.bibl.u-szeged.hu/12784
LEADER 01399nab a2200229 i 4500
001 acta12784
005 20220616101126.0
008 161015s2006 hu o 0|| eng d
022 |a 0324-721X 
040 |a SZTE Egyetemi Kiadványok Repozitórium  |b hun 
041 |a eng 
100 1 |a Szabó Zoltán 
245 1 0 |a ϵ-sparse representations   |h [elektronikus dokumentum] :  |b generalized sparse approximation and the equivalent family of SVM tasks /  |c  Szabó Zoltán 
260 |c 2006 
300 |a 605-614 
490 0 |a Acta cybernetica  |v 17 No. 3 
520 3 |a Relation between a family of generalized Support Vector Machine (SVM) problems and the novel ϵ-sparse representation is provided. In defining ϵ-sparse representations, we use a natural generalization of the classical ϵ-insensitive cost function for vectors. The insensitive parameter of the SVM problem is transformed into component-wise insensitivity and thus overall sparsification is replaced by component-wise sparsification. The connection between these two problems is built through the generalized Moore-Penrose inverse of the Gram matrix associated to the kernel. 
650 4 |a Természettudományok 
650 4 |a Számítás- és információtudomány 
695 |a Számítástechnika, Kibernetika 
700 0 1 |a Lőrincz András  |e aut 
856 4 0 |u http://acta.bibl.u-szeged.hu/12784/1/Lorincz_2006_ActaCybernetica.pdf  |z Dokumentum-elérés