Self-training is an important technique for solving semi-supervised learning problems. It leverages unlabeled data by generating pseudo-labels and combining them with a limited labeled dataset for training.
BibTex:
Before browse our site, please accept our cookies policy
