AutoNet-Generated Deep Layer-Wise Convex Networks for ECG Classification

The design of neural networks typically involves trial-and-error, a time-consuming process for obtaining an optimal architecture, even for experienced researchers. Additionally, it is widely accepted that loss functions of deep neural networks are generally non-convex with respect to the parameters...

Ausführliche Beschreibung

Bibliographische Detailangaben
Veröffentlicht in:IEEE transactions on pattern analysis and machine intelligence. - 1979. - 46(2024), 10 vom: 09. Sept., Seite 6542-6558
1. Verfasser: Shen, Yanting (VerfasserIn)
Weitere Verfasser: Lu, Lei, Zhu, Tingting, Wang, Xinshao, Clifton, Lei, Chen, Zhengming, Clarke, Robert, Clifton, David A
Format: Online-Aufsatz
Sprache:English
Veröffentlicht: 2024
Zugriff auf das übergeordnete Werk:IEEE transactions on pattern analysis and machine intelligence
Schlagworte:Journal Article