When deployed for risk-sensitive tasks, deep neural networks must be able to detect instances with labels from outside the distribution for which they were trained. In this paper we present a novel framework to benchmark the ability of image classifiers to detect class-out-of-distribution instances (i.e., instances whose true labels do not appear in the training distribution) at various levels of detection difficulty. We apply this technique to ImageNet, and benchmark 525 pretrained, publicly available, ImageNet-1k classifiers. The code for generating a benchmark for any ImageNet-1k classifier, along with the benchmarks prepared for the above-mentioned 525 models is available at https://github.com/mdabbah/COOD_benchmarking. The usefulness of the proposed framework and its advantage over alternative existing benchmarks is demonstrated by analyzing the results obtained for these models, which reveals numerous novel observations including: (1) knowledge distillation consistently improves class-out-of-distribution (C-OOD) detection performance; (2) a subset of ViTs performs better C-OOD detection than any other model; (3) the language--vision CLIP model achieves good zero-shot detection performance, with its best instance outperforming 96% of all other models evaluated; (4) accuracy and in-distribution ranking are positively correlated to C-OOD detection; and (5) we compare various confidence functions for C-OOD detection. Our companion paper, also published in ICLR 2023 (What Can We Learn From The Selective Prediction And Uncertainty Estimation Performance Of 523 Imagenet Classifiers), examines the uncertainty estimation performance (ranking, calibration, and selective prediction performance) of these classifiers in an in-distribution setting.

翻译：在风险敏感任务部署中，深度神经网络必须能够检测训练分布外标签的实例。本文提出了一种新颖的框架，用于在不同检测难度水平下评估图像分类器检测类外分布实例（即真实标签未出现在训练分布中的实例）的能力。我们将该方法应用于ImageNet，并对525个经过预训练、公开可用的ImageNet-1k分类器进行了基准测试。生成任意ImageNet-1k分类器基准测试的代码，以及为上述525个模型准备的基准测试结果，均可从https://github.com/mdabbah/COOD_benchmarking获取。通过分析这些模型的结果，我们展示了所提出框架的有效性及其相对于现有替代基准的优越性，并揭示了多项新发现，包括：(1) 知识蒸馏持续提升类外分布（C-OOD）检测性能；(2) 部分ViT模型在C-OOD检测方面优于所有其他模型；(3) 语言-视觉CLIP模型实现了良好的零样本检测性能，其最佳实例优于96%的其他评估模型；(4) 准确性与分布内排序与C-OOD检测呈正相关；(5) 我们比较了多种用于C-OOD检测的置信度函数。我们的姊妹论文（同样发表于ICLR 2023，题为《从523个ImageNet分类器的选择性预测与不确定性估计性能中我们能学到什么》）考察了这些分类器在分布内设置下的不确定性估计性能（排序、校准及选择性预测性能）。