We propose a novel semi-supervised learning (SSL) method that adopts selective training with pseudo labels. In our method, we generate hard pseudo-labels and also estimate their confidence, which represents how likely each pseudo-label is to be correct. Then, we explicitly select which pseudo-labeled data should be used to update the model. Specifically, assuming that loss on incorrectly pseudo-labeled data sensitively increase against data augmentation, we select the data corresponding to relatively small loss after applying data augmentation. The confidence is used not only for screening candidates of pseudo-labeled data to be selected but also for automatically deciding how many pseudo-labeled data should be selected within a mini-batch. Since accurate estimation of the confidence is crucial in our method, we also propose a new data augmentation method, called MixConf, that enables us to obtain confidence-calibrated models even when the number of training data is small. Experimental results with several benchmark datasets validate the advantage of our SSL method as well as MixConf.
翻译:我们提出一种新的半监督的学习方法,采用使用假标签的选择性培训。 在我们的方法中, 我们生成了硬假标签, 并估算了他们的信心, 这代表了每个假标签都正确的可能性。 然后, 我们明确选择了应该使用哪些假标签数据来更新模型。 具体地说, 我们假设错误的假标签数据损失会随着数据增强而敏感地增加, 我们选择了在应用数据增强后相对较小的损失的数据。 这种信任不仅用于筛选要选择的伪标签数据候选者, 而且还用于自动决定在微型批中选择多少伪标签数据。 由于准确估计信任对于我们的方法至关重要, 我们还提出了一种新的数据增强方法, 叫做 MixConf, 使我们能够获得信心调整模型, 即使培训数据的数量很小。 实验结果通过几个基准数据集验证了我们SL方法以及MixConf的优势。