MoCo (Momentum Contrast) framework

E1153668 UNEXPLORED

MoCo (Momentum Contrast) is a self-supervised learning framework for visual representation learning that uses a dynamic memory bank and momentum-updated encoder to enable effective contrastive learning on large-scale unlabeled data.

Try in SPARQL Jump to: Surface forms Referenced by

All labels observed (2)

Label	Occurrences
MoCo (Momentum Contrast) framework canonical	1
Momentum Contrast for Unsupervised Visual Representation Learning	1

Referenced by (2)

Full triples — surface form annotated when it differs from this entity's canonical label.

Kaiming He → knownFor → MoCo (Momentum Contrast) framework ⓘ

Kaiming He → notableWork → MoCo (Momentum Contrast) framework ⓘ

this entity surface form: Momentum Contrast for Unsupervised Visual Representation Learning