Variational Nested Dropout

Nested dropout is a variant of dropout operation that is able to order network parameters or features based on the pre-defined importance during training. It has been explored for: I. Constructing nested nets Cui et al. 2020, Cui et al. 2021: the nested nets are neural networks whose architectures c...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transactions on pattern analysis and machine intelligence 2023-08, Vol.45 (8), p.10519-10534
Hauptverfasser:	Cui, Yufei, Mao, Yu, Liu, Ziquan, Li, Qiao, Chan, Antoni B., Liu, Xue, Kuo, Tei-Wei, Xue, Chun Jason
Format:	Artikel
Sprache:	eng
Schlagworte:	Bayes methods Bayesian neural netowrk Coders Computational modeling Costs dropout Indexes Machine learning Mathematical models model compression Neural networks Parameters Representation learning Representations slimmable neural network Training Uncertainty uncertainty estimation variational autoencoder
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!