Scaling description of generalization with number of parameters in deep learning

Supervised deep learning involves the training of neural networks with a large number N of parameters. For large enough N, in the so-called over-parametrized regime, one can essentially fit the training data points. Sparsity-based arguments would suggest that the generalization error increases as N...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of statistical mechanics 2020-02, Vol.2020 (2), p.23401
Hauptverfasser:	Geiger, Mario, Jacot, Arthur, Spigler, Stefano, Gabriel, Franck, Sagun, Levent, d'Ascoli, Stéphane, Biroli, Giulio, Hongler, Clément, Wyart, Matthieu
Format:	Artikel
Sprache:	eng
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!