Contrastive estimation reveals topic posterior information to linear models

Tosh Christopher
Tosh Christopher
Cited by: 5|Views15

Abstract:

Contrastive learning is an approach to representation learning that utilizes naturally occurring similar and dissimilar pairs of data points to find useful embeddings of data. In the context of document classification under topic modeling assumptions, we prove that contrastive learning is capable of recovering a representation of docume...More

Code:

Data:

Full Text
Bibtex
Your rating :
0

 

Tags
Comments