Knowledge Distillation with Contrastive Inter-Class Relationship

Due to the high computational cost, the application of deep neural networks (DNNs) to the real-time tasks has been limited. A possible solution is to compress the size of the model so that the demand for computation resources can be decreased. A popular method is called knowledge distillation (KD)....

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of physics. Conference series 2021-02, Vol.1756 (1), p.12001
Hauptverfasser:	Yang, Chaoyi, Zeng, Jijun, Zhang, Jinbo
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial neural networks Distillation Knowledge management Machine learning Physics Teachers
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!