无需干净验证集，用带噪数据也能实现高效早停。
Noisy Early Stopping for Noisy Labels
- 直接监控与训练集同分布的带噪验证集准确率
- 在多个基准上达到近最优早停效果
- 适合标签噪声常见且无法获取干净验证集的场景
在标签含噪的数据集上训练神经网络分类器会显著增加过拟合风险。因此，在噪声环境下有效实施早停至关重要。理想情况下，早停依赖未受标签噪声污染的验证集来监控泛化性能。然而，获取无噪验证集成本高且困难。本研究发现，在多数典型学习环境中，无需干净验证集即可实现有效早停。通过监测与训练集同分布的带噪数据上的准确率，即可获得近最优结果。该方法称为「带噪早停」（NES），简化并降低了早停的实施成本。我们提供了该方法有效的理论条件分析，并在标准基准上使用常见损失函数实证验证了其稳健性能。
原文摘要 · Abstract (English)
Training neural network classifiers on datasets contaminated with noisy labels significantly increases the risk of overfitting. Thus, effectively implementing Early Stopping in noisy label environments is crucial. Under ideal circumstances, Early Stopping utilises a validation set uncorrupted by label noise to effectively monitor generalisation during training. However, obtaining a noise-free validation dataset can be costly and challenging to obtain. This study establishes that, in many typical learning environments, a noise-free validation set is not necessary for effective Early Stopping. Instead, near-optimal results can be achieved by monitoring accuracy on a noisy dataset - drawn from the same distribution as the noisy training set. Referred to as `Noisy Early Stopping' (NES), this method simplifies and reduces the cost of implementing Early Stopping. We provide theoretical insights into the conditions under which this method is effective and empirically demonstrate its robust performance across standard benchmarks using common loss functions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。