Universum Prescription: Regularization Using Unlabeled Data
release_qbg62yftnvfk5ctkubl6zj7hni
by
Xiang Zhang, Yann LeCun
2017 Volume 31
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 results in training deep convolutional networks for CIFAR-10, CIFAR-100, STL-10 and ImageNet datasets. A qualitative justification of these approaches using Rademacher complexity is presented. The effect of a regularization parameter — probability of sampling from unlabeled data — is also studied empirically.
In application/xml+jats
format
Archived Files and Locations
application/pdf 700.2 kB
file_5abroojmajf3dgwisgcbscyfj4
|
ojs.aaai.org (publisher) web.archive.org (webarchive) |
article-journal
Stage
published
Date 2017-02-13
access all versions, variants, and formats of this works (eg, pre-prints)
Crossref Metadata (via API)
Worldcat
SHERPA/RoMEO (journal policies)
wikidata.org
CORE.ac.uk
Semantic Scholar
Google Scholar