Not all data in a typical training set help with generalization; some samples
can be overly ambiguous or outrightly mislabeled. This paper introduces a new
method to identify such samples and mitigate their impact when training neural
networks. At the heart of our algorithm is the Area Under the Margin (AUM)
statistic, which exploits differences in the training dynamics of clean and
mislabeled samples. A simple procedure - adding an extra class populated with
purposefully mislabeled threshold samples - learns a AUM upper bound that
isolates mislabeled data. This approach consistently improves upon prior work
on synthetic and real-world datasets. On the WebVision50 classification task
our method removes 17% of training data, yielding a 1.6% (absolute) improvement
in test error. On CIFAR100 removing 13% of the data leads to a 1.2% drop in
error.

文章提出了一种新方法，使用区域下的边缘（AUM）统计量识别训练集中的错误标记数据，并通过添加一个带有特定错误标记样本的额外类来学习 AUM 上限以分离标记错误的数据。在 WebVision50 分类任务中，该方法去除了 17％的训练数据，产生了 1.6％（绝对）的测试误差改进，在 CIFAR100 上删除 13％的数据可导致 1.2％的误差下降。