DFEF: Diversify feature enhancement and fusion for online knowledge distillation

Traditional knowledge distillation relies on high‐capacity teacher models to supervise the training of compact student networks. To avoid the computational resource costs associated with pretraining high‐capacity teacher models, teacher‐free online knowledge distillation methods have achieved satisf...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Expert systems 2024-09, Vol.41 (9), p.n/a
Hauptverfasser:	Liang, Xingzhu, Zhang, Jian, Liu, Erhu, Fang, Xianjin
Format:	Artikel
Sprache:	eng
Schlagworte:	feature enhancement feature fusion Feature maps Knowledge Knowledge representation Modules online knowledge distillation Performance evaluation self‐distillation Teaching methods
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!