基于重构误差和典型惩罚的分布外检测
Out-of-Distribution Detection with Reconstruction Error and Typicality-based Penalty
摘要:
在实际应用中,离线检测对于实现安全可靠的运行至关重要。在基于似然的检测方法在高维空间中的失败被证明之后,基于\emph{typical set}的方法引起了人们的关注,然而,它们仍然没有取得令人满意的性能。通过介绍基于典型性的方法的失败情况开始,我们提出了一种新的重建误差为基础的方法,采用归一化流(NF)。我们进一步引入了一个基于典型性的惩罚,并将其纳入NF中的重建误差,我们提出了一种新的OOD检测方法,惩罚重建误差(PRE)。由于PRE检测到分布流形之外的测试输入,因此它可以有效地检测对抗性示例和OOD示例。我们通过使用自然图像数据集,CIFAR-10,TinyImageNet和ILSVRC 2012的评估来证明我们的方法的有效性。
英文摘要:
The task of out-of-distribution (OOD) detection is vital to realize safe and reliable operation for real-world applications. After the failure of likelihood-based detection in high dimensions had been shown, approaches based on the \emph{typical set} have been attracting attention; however, they still have not achieved satisfactory performance. Beginning by presenting the failure case of the typicality-based approach, we propose a new reconstruction error-based approach that employs normalizing flow (NF). We further introduce a typicality-based penalty, and by incorporating it into the reconstruction error in NF, we propose a new OOD detection method, penalized reconstruction error (PRE). Because the PRE detects test inputs that lie off the in-distribution manifold, it effectively detects adversarial examples as well as OOD examples. We show the effectiveness of our method through the evaluation using natural image datasets, CIFAR-10, TinyImageNet, and ILSVRC2012.
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。