Home Page

Papers

Submissions

News

Editorial Board

Proceedings (PMLR)

Transactions (TMLR)

Open Source Software

Search

Statistics

Login

Frequently Asked Questions

Contact Us



RSS Feed

Contrastive Estimation Reveals Topic Posterior Information to Linear Models

Christopher Tosh, Akshay Krishnamurthy, Daniel Hsu; 22(281):1−31, 2021.

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 documents that reveals their underlying topic posterior information to linear models. We apply this procedure in a semi-supervised setup and demonstrate empirically that linear classifiers trained on these representations perform well in document classification tasks with very few training examples.

[abs][pdf][bib]       
© JMLR 2021. (edit, beta)