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:当用于风险敏感任务时,深心神经网络必须能够检测到从外部分发的标签显示其经过培训的分布图象的情况。在本文中,我们提出了一个新的框架,以衡量图像分类者在检测困难的不同级别检测分类(即真正标签在培训分布中没有出现的实例)的能力。我们将这一技术应用于图像网络,并将525种预先培训的、可公开获取的图像Net-1k分类器的基准用于基准525。为任何图像Net-1k分类器创建基准的代码,以及为上述525模型制定的基准。在https://github.com/mdabbah/COOD_benchmarketing中,我们提供了一个新的框架的有用性及其相对于现有替代基准的优势,通过分析这些模型获得的结果,我们展示了许多新的观察,包括:(1) 知识蒸馏不断改进分类、可公开获取的、可公开获取的IMD-23的检测性;(2) ViTs的所有部分对C-OOO值的检测比任何其他模型更好;(3) 从语言-O级的检测结果模型到我们的C-C-Silder-Silal-dealation Stalation Oder Oder supalation Exeral deal deal demodealation ex ex ex exalationalationalational deal deal deal deal deal deal deal ex ex a ex a ex a ex ex ex ex ex ex ex ex ex ex ex ex exal deal deal deal demoal deal demomental deal deal demomental dealdal deal deal deal deal deal demoment a exment a ex a ex a ex a ex a ex a exmental deal deal deal dealds a ex a ex a ex a ex a exal deal deal deald sal deal deal deal deal deal deal deal deal de ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex a ex