Monte Carlo averaging for uncertainty estimation in neural networks

Although convolutional neural networks (CNNs) are widely used in modern classifiers, they are affected by overfitting and lack robustness leading to overconfident false predictions (FPs). By preventing FPs, certain consequences (such as accidents and financial losses) can be avoided and the use of C...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Journal of physics. Conference series 2023-05, Vol.2506 (1), p.12004
Hauptverfasser:	Tassi, Cedrique Rovile Njieutcheu, Börner, Anko, Triebel, Rudolph
Format:	Artikel
Sprache:	eng
Schlagworte:	Artificial neural networks confidence calibration Convolutional neural network (CNN) ensemble mixture of Monte Carlo dropout (MMCD) Monte Carlo averaging (MCA) Monte Carlo dropout (MCD) Neural networks Physics separating true predictions (TPs) and false predictions (FPs) Uncertainty
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!