Glossary
What is InfoNCE? Definition and meaning
A loss that raises the score of a true pair relative to a set of negatives, a lower bound on mutual information.
Definition
InfoNCE: A loss that raises the score of a true pair relative to a set of negatives, a lower bound on mutual information.
In short: The standard contrastive loss.
When it is used
Contrastive pre-training of embeddings.
Learn more
The full guide to InfoNCE explains how it works, when to use it and how it compares with the alternatives, with examples. The glossary lists every term on the site from A to Z.
Previous term: Industry 4.0 · Next term: Informatica Axon