We present Meta Pseudo Labels, a semi-supervised learning method that achieves a new state-of-the-art top-1 accuracy of 90.2% on ImageNet, which is 1.6% better than the existing state-of-the-art. Like Pseudo Labels, Meta Pseudo Labels has a teacher network to generate pseudo labels on unlabeled data to teach a student network. However, unlike Pseudo Labels where the teacher is fixed, the teacher in Meta Pseudo Labels is constantly adapted by the feedback of the student's performance on the labeled dataset. As a result, the teacher generates better pseudo labels to teach the student. Our code will be available at https://github.com/google-research/google-research/tree/master/meta_pseudo_labels.

翻译：我们展示了Meta Pseudo Labels, 这是一种半监督的学习方法,在图像网络上实现了90.2%的最先进的最高一级精确度为90.2%,比现有最先进的精确度高1.6%。像Pseudo Labels一样,Meta Pseudo Labels有一个教师网络,在未贴标签的数据上生成假标签,用于教授学生网络。然而,与教师固定的Pseudo Labels不同,Meta Pseudo Labels的教师不断因学生在标签数据集上的表现反馈而适应。结果,该教师制作了更好的假标签,教学生。我们的代码将在https://github.com/google-research/tree/master/meta_psedodo_lables上查阅。