Lookaround Optimizer: \(k\) steps around, 1 step average

Weight Average (WA) is an active research topic due to its simplicity in ensembling deep networks and the effectiveness in promoting generalization. Existing weight average approaches, however, are often carried out along only one training trajectory in a post-hoc manner (i.e., the weights are avera...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2023-11
Hauptverfasser:	Zhang, Jiangtao, Liu, Shunyu, Song, Jie, Zhu, Tongtian, Xu, Zhengqi, Song, Mingli
Format:	Artikel
Sprache:	eng
Schlagworte:	Data augmentation Effectiveness Networks Training
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!