你能做到多低？为无监督异常检测挖掘典型分布内样本
How Low Can You Go? Surfacing Prototypical In-Distribution Samples for Unsupervised Anomaly Detection
AI总结：
针对无监督异常检测依赖大规模训练数据的假设，本文提出无监督识别典型分布内样本的方法，仅用25个样本即可在多个基准中匹敌甚至超过全量训练。
摘要：
无监督异常检测（UAD）通过仅在无标签分布内数据上训练并将离群值检测为异常，减轻了大量标注工作。通常普遍认为，大型训练数据集能够训练出性能更高的UAD模型。然而，在本工作中，我们表明使用极少训练样本的UAD已经可以匹敌——在某些情况下甚至超过——使用整个训练数据集训练的性能。基于这一发现，我们提出一种无监督方法，用于可靠识别典型样本，以进一步提升UAD性能。我们在来自计算机视觉、工业缺陷检测和医学领域的七个不同成熟UAD基准上展示了该方法的效用。仅使用25个选定样本，我们甚至在这些基准的25/67个类别中超过了完整训练的性能。此外，我们表明，由所提方法识别出的典型分布内样本能够在不同模型和数据集之间良好泛化，并且观察其样本选择准则可以成功人工选择高性能样本的小子集。我们的代码可在https://anonymous.4open.science/r/uad_prototypical_samples/获取。
英文摘要：
Unsupervised anomaly detection (UAD) alleviates large labeling efforts by training exclusively on unlabeled in-distribution data and detecting outliers as anomalies. Generally, the assumption prevails that large training datasets allow the training of higher-performing UAD models. However, in this work, we show that UAD with extremely few training samples can already match -- and in some cases even surpass -- the performance of training with the whole training dataset. Building upon this finding, we propose an unsupervised method to reliably identify prototypical samples to further boost UAD performance. We demonstrate the utility of our method on seven different established UAD benchmarks from computer vision, industrial defect detection, and medicine. With just 25 selected samples, we even exceed the performance of full training in $25/67$ categories in these benchmarks. Additionally, we show that the prototypical in-distribution samples identified by our proposed method generalize well across models and datasets and that observing their sample selection criteria allows for a successful manual selection of small subsets of high-performing samples. Our code is available at https://anonymous.4open.science/r/uad_prototypical_samples/
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
