A New Data Selection Principle for Semi-Supervised Incremental Learning

Current semi-supervised incremental learning approaches select unlabeled examples with predicted high confidence for model re-training. We show that for many applications this data selection strategy is not correct. This is because the confidence score is primarily a metric to measure the classifica...

Ausführliche Beschreibung

Bibliographische Detailangaben

Hauptverfasser:	Rong Zhang, Rudnicky, A.I.
Format:	Tagungsbericht
Sprache:	eng
Schlagworte:	Bridges Degradation Image retrieval Information analysis Information retrieval Particle measurements Predictive models Semisupervised learning Speech recognition Variable structure systems
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!