In this work we consider univariate global optimization tools for a better detection of the kernel shape parameter in radial basis function interpolation methods. In particular, we use efficient global optimization techniques with pessimistic and optimistic improvement in combination with a leave-one-out cross validation scheme. Numerical results point out some enhancements in the performance of the new methods.
Global optimization tools for shape parameter detection in RBF and RBF-PU methods
Cavoretto, Roberto
First
;Rossi, Alessandra De;
2025-01-01
Abstract
In this work we consider univariate global optimization tools for a better detection of the kernel shape parameter in radial basis function interpolation methods. In particular, we use efficient global optimization techniques with pessimistic and optimistic improvement in combination with a leave-one-out cross validation scheme. Numerical results point out some enhancements in the performance of the new methods.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
P39.pdf
Accesso aperto
Tipo di file:
PDF EDITORIALE
Dimensione
705.15 kB
Formato
Adobe PDF
|
705.15 kB | Adobe PDF | Visualizza/Apri |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
