用类别自编码器比值检测图像分类错误，能精准定位标签错误和困难样本。
Class-wise Autoencoders Measure Classification Difficulty And Detect Label Mistakes
- 通过单类自编码器的重建误差比，量化分类难度
- 在19个数据集上与SOTA模型错误率高度相关，准确率超现有方法
- 适合数据清洗、标注质量检查，尤其对噪声数据有效
我们提出一种基于单类自编码器重建误差比的新框架，用于分析分类数据集。该框架可在样本、类别和整个数据集层面高效表征数据。定义重建误差比（RER）来探测分类难度，并将其分解为有限样本量影响和贝叶斯误差与决策边界复杂性两部分。在19个主流视觉数据集上的系统研究表明，RER与当前最优分类模型的错误率强相关。将样本级难度视为标签错误度量，RER在对称与非对称标签噪声下的困难数据集上，于误标检测任务中达到最先进性能。代码已公开于 https://github.com/voxel51/reconstruction-error-ratios。
原文摘要 · Abstract (English)
We introduce a new framework for analyzing classification datasets based on the ratios of reconstruction errors between autoencoders trained on individual classes. This analysis framework enables efficient characterization of datasets on the sample, class, and entire dataset levels. We define reconstruction error ratios (RERs) that probe classification difficulty and allow its decomposition into (1) finite sample size and (2) Bayes error and decision-boundary complexity. Through systematic study across 19 popular visual datasets, we find that our RER-based dataset difficulty probe strongly correlates with error rate for state-of-the-art (SOTA) classification models. By interpreting sample-level classification difficulty as a label mistakenness score, we further find that RERs achieve SOTA performance on mislabel detection tasks on hard datasets under symmetric and asymmetric label noise. Our code is publicly available at https://github.com/voxel51/reconstruction-error-ratios.
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。