用虚拟异常原型改善模型对未知数据的误判问题
Out-of-Distribution Detection with Prototypical Outlier Proxy
- 引入虚拟异常原型重塑分类边界，固定分类器权重增强稳定性
- 在多个数据集上将误报率降低超5%，最高达7.7%
- 训练和推理速度比同类方法快7倍以上，适合实际部署
分布外（OOD）检测是深度学习模型野外部署的关键任务。现有方法常因过度自信于未见数据而失效，部分通过真实或合成异常样本缓解，但计算成本高且偏向特定异常特征。本文提出简单有效的框架POP，引入虚拟OOD原型以调整ID与OOD间的决策边界。具体地，将可学习分类器转为固定形式，并添加一组原型权重向量；设计分层相似性边界损失，按误分类程度动态施加惩罚。大量实验表明，POP在CIFAR-10、CIFAR-100、ImageNet-200上分别较次优方法降低平均FPR95 7.70%、6.30%、5.42%。相比依赖异常生成的NPOS方法，POP训练速度快7.2倍，推理快19.5倍。代码已开源。
原文摘要 · Abstract (English)
Out-of-distribution (OOD) detection is a crucial task for deploying deep learning models in the wild. One of the major challenges is that well-trained deep models tend to perform over-confidence on unseen test data. Recent research attempts to leverage real or synthetic outliers to mitigate the issue, which may significantly increase computational costs and be biased toward specific outlier characteristics. In this paper, we propose a simple yet effective framework, Prototypical Outlier Proxy (POP), which introduces virtual OOD prototypes to reshape the decision boundaries between ID and OOD data. Specifically, we transform the learnable classifier into a fixed one and augment it with a set of prototypical weight vectors. Then, we introduce a hierarchical similarity boundary loss to impose adaptive penalties depending on the degree of misclassification. Extensive experiments across various benchmarks demonstrate the effectiveness of POP. Notably, POP achieves average FPR95 reductions of 7.70%, 6.30%, and 5.42% over the second-best methods on CIFAR-10, CIFAR-100, and ImageNet-200, respectively. Moreover, compared to the recent method NPOS, which relies on outlier synthesis, POP trains 7.2X faster and performs inference 19.5X faster. The source code is available at: https://github.com/gmr523/pop.
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。