Semi-supervised learning (SSL) algorithm is a setup built upon a realistic assumption that access to a large amount of labeled data is tough. In this study, we present a generalized framework, named SCAR, standing for Selecting Clean samples with Adversarial Robustness, for improving the performance of recent SSL algorithms. By adversarially attacking pre-trained models with semi-supervision, our framework shows substantial advances in classifying images. We introduce how adversarial attacks successfully select high-confident unlabeled data to be labeled with current predictions. On CIFAR10, three recent SSL algorithms with SCAR result in significantly improved image classification.
@article{arxiv.2308.04018,
title = {Improving Performance of Semi-Supervised Learning by Adversarial Attacks},
author = {Dongyoon Yang and Kunwoong Kim and Yongdai Kim},
journal= {arXiv preprint arXiv:2308.04018},
year = {2023}
}