Semi-supervised Learning (SSL) has witnessed great success owing to the impressive performances brought by various methods based on pseudo labeling and consistency regularization. However, we argue that existing methods might fail to utilize the unlabeled data more effectively since they either use a pre-defined / fixed threshold or an ad-hoc threshold adjusting scheme, resulting in inferior performance and slow convergence. We first analyze a motivating example to obtain intuitions on the relationship between the desirable threshold and model's learning status. Based on the analysis, we hence propose FreeMatch to adjust the confidence threshold in a self-adaptive manner according to the model's learning status. We further introduce a self-adaptive class fairness regularization penalty to encourage the model for diverse predictions during the early training stage. Extensive experiments indicate the superiority of FreeMatch especially when the labeled data are extremely rare. FreeMatch achieves 5.78%, 13.59%, and 1.28% error rate reduction over the latest state-of-the-art method FlexMatch on CIFAR-10 with 1 label per class, STL-10 with 4 labels per class, and ImageNet with 100 labels per class, respectively. Moreover, FreeMatch can also boost the performance of imbalanced SSL. The codes can be found at https://github.com/microsoft/Semi-supervised-learning.

翻译：半监督学习（SSL）因基于伪标签和一致性正则化的多种方法所带来的卓越表现而取得了巨大成功。然而，我们认为现有方法可能无法更有效地利用未标记数据，因为它们要么使用预定义/固定阈值，要么采用临时调整方案，导致性能较差且收敛缓慢。我们首先分析了一个激励性示例，以获取关于理想阈值与模型学习状态之间关系的直觉。基于此分析，我们提出了FreeMatch，根据模型的学习状态以自适应方式调整置信度阈值。我们进一步引入了一种自适应的类别公平性正则化惩罚项，以在早期训练阶段鼓励模型进行多样化预测。大量实验表明，FreeMatch在标记数据极其稀缺的情况下尤其具有优势。与最新的最先进方法FlexMatch相比，FreeMatch在CIFAR-10（每类1个标签）、STL-10（每类4个标签）和ImageNet（每类100个标签）上分别实现了5.78%、13.59%和1.28%的错误率降低。此外，FreeMatch还能提升不平衡半监督学习的性能。代码可从 https://github.com/microsoft/Semi-supervised-learning 获取。