All labels are not created equal: enhancing semi-supervision via label grouping and co-training

Islam Nassar, Samitha Herath, Ehsan Abbasnejad , Wray Buntine, Reza Haffari

Research output: Chapter in Book/Report/Conference proceedingConference PaperResearchpeer-review

12 Citations (Scopus)

Abstract

Pseudo-labeling is a key component in semi-supervised learning (SSL). It relies on iteratively using the model to generate artificial labels for the unlabeled data to train against. A common property among its various methods is that they only rely on the model's prediction to make labeling decisions without considering any prior knowledge about the visual similarity among the classes. In this paper, we demonstrate that this degrades the quality of pseudo-labeling as it poorly represents visually similar classes in the pool of pseudo-labeled data. We propose SemCo, a method which leverages label semantics and co-training to address this problem. We train two classifiers with two different views of the class labels: one classifier uses the one-hot view of the labels and disregards any potential similarity among the classes, while the other uses a distributed view of the labels and groups potentially similar classes together. We then co-train the two classifiers to learn based on their disagreements. We show that our method achieves state-of-the-art performance across various SSL tasks including 5.6% accuracy improvement on Mini-ImageNet dataset with 1000 labeled examples. We also show that our method requires smaller batch size and fewer training iterations to reach its best performance. We make our code available at https://github.com/islam-nassar/semco.

Original languageEnglish
Title of host publicationProceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
EditorsMargaux Masson-Forsythe, Eric Mortensen
Place of PublicationPiscataway NJ USA
PublisherIEEE, Institute of Electrical and Electronics Engineers
Pages7237-7246
Number of pages10
ISBN (Electronic)9781665445092
ISBN (Print)9781665445108
DOIs
Publication statusPublished - 2021
EventIEEE Conference on Computer Vision and Pattern Recognition 2021 - Online, United States of America
Duration: 19 Jun 202125 Jun 2021
https://cvpr2021.thecvf.com/ (Website)
https://ieeexplore.ieee.org/xpl/conhome/9577055/proceeding (Proceedings)

Publication series

NameProceedings 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPR 2021
PublisherThe Institute of Electrical and Electronics Engineers, Inc.
ISSN (Print)2575-7075
ISSN (Electronic)2575-7075

Conference

ConferenceIEEE Conference on Computer Vision and Pattern Recognition 2021
Abbreviated titleCVPR 2021
Country/TerritoryUnited States of America
Period19/06/2125/06/21
Internet address

Cite this