针对二分类数据不平衡问题，改进对比学习以提升模型性能。
A Tale of Two Classes: Adapting Supervised Contrastive Learning to Binary Imbalanced Datasets
- 提出新方法优化不平衡数据下的对比学习表示空间。
- 在7个数据集上使分类准确率提升最高达35%。
- 适合医疗影像等真实场景中类别不均衡的问题。
监督对比学习（SupCon）在多类平衡数据集上表现优异，但在长尾分布数据上效果下降。本文在七个自然与医学图像的二分类数据集上发现，随着类别不平衡程度增加，SupCon性能显著降低。为揭示原因，提出两个新度量指标，通过分析局部邻域中的类别分布，发现了传统指标无法察觉的表示空间结构性缺陷。基于此，提出两种专为二分类不平衡数据设计的监督对比学习策略，有效改善表示空间结构，使下游分类准确率相比标准SupCon最高提升35%。代码已开源。
原文摘要 · Abstract (English)
Supervised contrastive learning (SupCon) has proven to be a powerful alternative to the standard cross-entropy loss for classification of multi-class balanced datasets. However, it struggles to learn well-conditioned representations of datasets with long-tailed class distributions. This problem is potentially exacerbated for binary imbalanced distributions, which are commonly encountered during many real-world problems such as medical diagnosis. In experiments on seven binary datasets of natural and medical images, we show that the performance of SupCon decreases with increasing class imbalance. To substantiate these findings, we introduce two novel metrics that evaluate the quality of the learned representation space. By measuring the class distribution in local neighborhoods, we are able to uncover structural deficiencies of the representation space that classical metrics cannot detect. Informed by these insights, we propose two new supervised contrastive learning strategies tailored to binary imbalanced datasets that improve the structure of the representation space and increase downstream classification accuracy over standard SupCon by up to 35%. We make our code available.
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。