Description
Meta Pseudo Labels is a semi-supervised learning method that uses a teacher network to generate pseudo labels on unlabeled data to teach a student network. The teacher receives feedback from the student to inform the teacher to generate better pseudo labels. This feedback signal is used as a reward to train the teacher throughout the course of the student’s learning.
Papers Using This Method
The Efficiency of Pre-training with Objective Masking in Pseudo Labeling for Semi-Supervised Text Classification2025-05-10Meta Co-Training: Two Views are Better than One2023-11-29Self Meta Pseudo Labels: Meta Pseudo Labels Without The Teacher2022-12-27Dense Interspecies Face Embedding2022-11-28Meta Pseudo Labels2020-03-23