用新检验方法融合对比学习得分，提升未知异常检测效果
Score Combining for Contrastive OOD Detection
- 基于广义似然比检验融合多个对比学习得分
- 在多个数据集上超越现有最优方法，尤其在留一类别实验中表现突出
- 适合需要高可靠性异常检测的工业场景或安全应用
在分布外（OOD）检测任务中，需判断测试样本是否来自已知的内联分布。本文聚焦于仅由训练数据定义内联分布、无额外新颖性先验知识的情况，即所谓的新颖性检测、一类分类和无监督异常检测。当前文献表明，对比学习技术在该任务中处于领先水平。本文通过零假设检验框架，特别是提出一种新的广义似然比检验（GLRT），对多个对比学习模型的得分进行融合。实验表明，所提GLRT方法在CIFAR-10、SVHN、LSUN、ImageNet和CIFAR-100等数据集上的数据集间对比实验，以及在CIFAR-10上的留一类别实验中，均优于现有最优的CSI与SupCSI方法。同时，相较于Fisher、Bonferroni、Simes、Benjamini-Hochberg及Stouffer等传统得分融合策略，本方法也表现出更优性能。
原文摘要 · Abstract (English)
In out-of-distribution (OOD) detection, one is asked to classify whether a test sample comes from a known inlier distribution or not. We focus on the case where the inlier distribution is defined by a training dataset and there exists no additional knowledge about the novelties that one is likely to encounter. This problem is also referred to as novelty detection, one-class classification, and unsupervised anomaly detection. The current literature suggests that contrastive learning techniques are state-of-the-art for OOD detection. We aim to improve on those techniques by combining/ensembling their scores using the framework of null hypothesis testing and, in particular, a novel generalized likelihood ratio test (GLRT). We demonstrate that our proposed GLRT-based technique outperforms the state-of-the-art CSI and SupCSI techniques from Tack et al. 2020 in dataset-vs-dataset experiments with CIFAR-10, SVHN, LSUN, ImageNet, and CIFAR-100, as well as leave-one-class-out experiments with CIFAR-10. We also demonstrate that our GLRT outperforms the score-combining methods of Fisher, Bonferroni, Simes, Benjamini-Hochwald, and Stouffer in our application.
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。