Milking CowMask for Semi-Supervised Image Classification

Understand

Consistency regularization is a technique for semi-supervised learning that underlies a number of strong results for classification with few labeled data.

• It works by encouraging a learned model to be robust to perturbations on unlabeled data.
• Here, we present a novel mask-based augmentation method called CowMask.
• Using it to provide perturbations for semi-supervised consistency regularization, we achieve a state-of-the-art result on ImageNet with 10% labeled data, with a top-5 error of 8.76% and top-1 error of 26.06%.