We deal with the problem of computing near G-optimal compressed designs for high-degree polynomial regression on fine discretizations of 2d and 3d regions of arbitrary shape. The key tool is Tchakaloff-like compression of discrete probability measures, via an improved version of the Lawson-Hanson NNLS solver for the corresponding full and large-scale underdetermined moment system, that can have for example a size order of 10ˆ3 (basis polynomials) x 10ˆ4 (nodes).
Accelerating the Lawson-Hanson NNLS solver for large-scale Tchakaloff regression designs
Dessole, M;Marcuzzi, F;Vianello, M
2020
Abstract
We deal with the problem of computing near G-optimal compressed designs for high-degree polynomial regression on fine discretizations of 2d and 3d regions of arbitrary shape. The key tool is Tchakaloff-like compression of discrete probability measures, via an improved version of the Lawson-Hanson NNLS solver for the corresponding full and large-scale underdetermined moment system, that can have for example a size order of 10ˆ3 (basis polynomials) x 10ˆ4 (nodes).
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
DessoleMarcuzziVianello_2020_ALH.pdf
accesso aperto
Tipologia:
Published (publisher's version)
Licenza:
Accesso gratuito
Dimensione
1.02 MB
Formato
Adobe PDF
|
1.02 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
