Open access
Aug 2026
The More the Merrier: The Effects of Data Volume and Transfer Learning on Creating Generalizable Deep Learning PAM Detectors
While trained models saw a drop in performance when tested on a new dataset, those pretrained on large diverse datasets outperformed limited target‐specific training even when deployment conditions were not represented during training, confirming training diversity improves transferability.
Laia Garrobé Fonollosa, D. Gillespie, L. Stanković et al.
· Marine mammal science · 0 citations