Operation is the Hardest Teacher: Estimating DNN Accuracy Looking for Mispredictions

Deep Neural Networks (DNN) are typically tested for accuracy relying on a set of unlabelled real world data (operational dataset), from which a subset is selected, manually labelled and used as test suite. This subset is required to be small (due to manual labelling cost) yet to faithfully represent...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	Proceedings / International Conference on Software Engineering S. 348 - 358
Hauptverfasser:	Guerriero, Antonio, Pietrantuono, Roberto, Russo, Stefano
Format:	Tagungsbericht
Sprache:	Englisch
Veröffentlicht:	IEEE 01.05.2021
Schlagworte:	Adaptation models Artificial Neural Networks Computer bugs Labeling Neural networks Software Software engineering Software testing Testing
ISBN:	1665402962, 9781665402965
ISSN:	1558-1225
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!