Random Field Augmentations for Self-Supervised Representation Learning

Self-supervised representation learning is heavily dependent on data augmentations to specify the invariances encoded in representations. Previous work has shown that applying diverse data augmentations is crucial to downstream performance, but augmentation techniques remain under-explored. In this...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2023-11
Hauptverfasser:	Mansfield, Philip Andrew, Afkanpour, Arash, Morningstar, Warren Richard, Singhal, Karan
Format:	Artikel
Sprache:	eng
Schlagworte:	Classification Color Continuity (mathematics) Data augmentation Fields (mathematics) Machine learning Parameters Pixels Representations Transformations
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!