They are Not Completely Useless: Towards Recycling Transferable Unlabeled Data for Class-Mismatched Semi-Supervised Learning. (arXiv:2011.13529v4 [cs.LG] UPDATED)

Semi-Supervised Learning (SSL) with mismatched classes deals with the problem
that the classes-of-interests in the limited labeled data is only a subset of
the classes in massive unlabeled data. As a result, the classes only possessed
by the unlabeled data may mislead the classifier training and thus hindering
the realistic landing of various SSL methods. To solve this problem, existing
methods usually divide unlabeled data to in-distribution (ID) data and
out-of-distribution (OOD) data, and directly discard or weaken the OOD …

arxiv data learning recycling semi-supervised semi-supervised learning supervised learning