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

do-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 la-
beling 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 dif-
ferent 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 includ-
ing 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.