Universum Prescription: Regularization using Unlabeled Data

national conference on artificial intelligence, 2017.

Cited by: 9|Bibtex|Views111
EI
Other Links: academic.microsoft.com|dblp.uni-trier.de|arxiv.org

Abstract:

This paper shows that simply prescribing "none of the above" labels to unlabeled data has a beneficial regularization effect to supervised learning. We call it universum prescription by the fact that the prescribed labels cannot be one of the supervised labels. In spite of its simplicity, universum prescription obtained competitive resu...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments