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 language | English |
---|---|
Title of host publication | Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021 |
Editors | Margaux Masson-Forsythe, Eric Mortensen |
Place of Publication | Piscataway NJ USA |
Publisher | IEEE, Institute of Electrical and Electronics Engineers |
Pages | 7237-7246 |
Number of pages | 10 |
ISBN (Electronic) | 9781665445092 |
ISBN (Print) | 9781665445108 |
DOIs | |
Publication status | Published - 2021 |
Event | IEEE Conference on Computer Vision and Pattern Recognition 2021 - Online, United States of America Duration: 19 Jun 2021 → 25 Jun 2021 https://cvpr2021.thecvf.com/ (Website) https://ieeexplore.ieee.org/xpl/conhome/9577055/proceeding (Proceedings) |
Publication series
Name | Proceedings 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition CVPR 2021 |
---|---|
Publisher | The Institute of Electrical and Electronics Engineers, Inc. |
ISSN (Print) | 2575-7075 |
ISSN (Electronic) | 2575-7075 |
Conference
Conference | IEEE Conference on Computer Vision and Pattern Recognition 2021 |
---|---|
Abbreviated title | CVPR 2021 |
Country/Territory | United States of America |
Period | 19/06/21 → 25/06/21 |
Internet address |
|