In this paper, we investigate neural models based on graph random features. In particular, we aim to understand when over-parameterization, namely generating more features than the ones necessary to interpolate, may be beneficial for the generalization of the resulting models. Exploiting the algorithmic stability framework and based on empirical evidences from several commonly adopted graph datasets, we will shed some light on this issue.
An Empirical Study of Over-Parameterized Neural Models based on Graph Random Features
Nicolo Navarin;Luca Pasa;Alessandro Sperduti
2023
Abstract
In this paper, we investigate neural models based on graph random features. In particular, we aim to understand when over-parameterization, namely generating more features than the ones necessary to interpolate, may be beneficial for the generalization of the resulting models. Exploiting the algorithmic stability framework and based on empirical evidences from several commonly adopted graph datasets, we will shed some light on this issue.
File in questo prodotto:
| File | Dimensione | Formato | |
|---|---|---|---|
|
2023 - An Empirical Study of Over-Parameterized Neural Models based on Graph Random Features.pdf
accesso aperto
Tipologia:
Published (Publisher's Version of Record)
Licenza:
Accesso libero
Dimensione
1.76 MB
Formato
Adobe PDF
|
1.76 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
