arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2303.13148cs.CV
收藏

校准的基于通用表征的分布外检测

Calibrated Out-of-Distribution Detection with a Generic Representation

Czech Technical University in Prague(布拉格捷克理工大学);Toyota Motor Europe(丰田汽车欧洲公司)

Tomas Vojir, Jan Sochman, Rahaf Aljundi, Jiri Matas

AI总结:

提出GROOD方法,利用通用预训练表征,将OOD检测视为Neyman-Pearson任务,实现校准良好的分数,在多个基准上达到最先进性能,且训练简单、超参数少。

摘要:

分布外检测是视觉模型在实际部署中的一个常见问题,解决该问题对于安全关键型应用至关重要。现有的大多数OOD检测解决方案专注于提高仅在分布内(ID)数据上训练的分类模型的OOD鲁棒性。在本工作中,我们采取不同的方法,提出利用通用预训练表征。我们提出了一种新颖的OOD方法,称为GROOD,该方法将OOD检测表述为具有良好校准分数的Neyman-Pearson任务,并在使用良好通用表征的前提下取得了优异的性能。将GROOD适应于特定问题仅需一个简单的训练过程。该方法简单、通用、高效、校准良好,且仅有少量超参数。该方法在多个OOD基准上达到了最先进的性能,并在其中几个基准上接近完美性能。源代码可在https://github.com/vojirt/GROOD获取。

英文摘要:

Out-of-distribution detection is a common issue in deploying vision models in practice and solving it is an essential building block in safety critical applications. Most of the existing OOD detection solutions focus on improving the OOD robustness of a classification model trained exclusively on in-distribution (ID) data. In this work, we take a different approach and propose to leverage generic pre-trained representation. We propose a novel OOD method, called GROOD, that formulates the OOD detection as a Neyman-Pearson task with well calibrated scores and which achieves excellent performance, predicated by the use of a good generic representation. Only a trivial training process is required for adapting GROOD to a particular problem. The method is simple, general, efficient, calibrated and with only a few hyper-parameters. The method achieves state-of-the-art performance on a number of OOD benchmarks, reaching near perfect performance on several of them. The source code is available at https://github.com/vojirt/GROOD.

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。

↑