MixMatch: A Holistic Approach to Semi-Supervised Learning
Abstract
Semi-supervised learning has proven to be a powerful paradigm for leveraging unlabeled data to mitigate the reliance on large labeled datasets. In this work, we unify the current dominant approaches for semi-supervised learning to produce a new algorithm, MixMatch, that works by guessing low-entropy labels for data-augmented unlabeled examples and mixing labeled and unlabeled data using MixUp. We show that MixMatch obtains state-of-the-art results by a large margin across many datasets and labeled data amounts. For example, on CIFAR-10 with 250 labels, we reduce error rate by a factor of 4 (from 38% to 11%) and by a factor of 2 on STL-10. We also demonstrate how MixMatch can help achieve a dramatically better accuracy-privacy trade-off for differential privacy. Finally, we perform an ablation study to tease apart which components of MixMatch are most important for its success.
Cite
@article{arxiv.1905.02249,
title = {MixMatch: A Holistic Approach to Semi-Supervised Learning},
author = {David Berthelot and Nicholas Carlini and Ian Goodfellow and Nicolas Papernot and Avital Oliver and Colin Raffel},
journal= {arXiv preprint arXiv:1905.02249},
year = {2019}
}