Milking CowMask for Semi-Supervised Image Classification
This addresses the problem of reducing labeled data requirements for image classification, with broad applicability across datasets, though it is incremental as it builds on existing consistency regularization techniques.
The paper tackles semi-supervised image classification by introducing CowMask, a novel mask-based augmentation method for consistency regularization, achieving state-of-the-art results on ImageNet with 10% labeled data, including a top-5 error of 8.76% and top-1 error of 26.06%.
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%. Moreover, we do so with a method that is much simpler than many alternatives. We further investigate the behavior of CowMask for semi-supervised learning by running many smaller scale experiments on the SVHN, CIFAR-10 and CIFAR-100 data sets, where we achieve results competitive with the state of the art, indicating that CowMask is widely applicable. We open source our code at https://github.com/google-research/google-research/tree/master/milking_cowmask