Unlabeled PCA-shuffling initialization for convolutional neural networks

In order to obtain prominent recognition accuracy convolutional neural networks (CNNs) need large amounts of labeled data to initialize network parameters. However, there exist two open problems, i.e., the uncertainties of the initialized effects and the limited labeled data To address the problems,...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Applied intelligence (Dordrecht, Netherlands) Netherlands), 2018-12, Vol.48 (12), p.4565-4576
Hauptverfasser:	Ou, Jun, Li, Yujian, Shen, Chengkai
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial Intelligence Artificial neural networks Computer Science Machines Manufacturing Mechanical Engineering Neural networks Parameter uncertainty Principal components analysis Processes Recognition Sampling
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!