In this work, we study the out-of-distribution (OOD) detection problem through the use of the feature space of a pre-trained deep classifier. We show that learning the density of in-distribution (ID) features with an energy-based models (EBM) leads to competitive detection results. However, we found that the non-mixing of MCMC sampling during the EBM's training undermines its detection performance. To overcome this an energy-based correction of a mixture of class-conditional Gaussian distributions. We obtains favorable results when compared to a strong baseline like the KNN detector on the CIFAR-10/CIFAR-100 OOD detection benchmarks.

翻译：本研究通过利用预训练深度分类器的特征空间，探讨了分布外（OOD）检测问题。我们证明，使用基于能量的模型（EBM）学习分布内（ID）特征的密度，能够获得具有竞争力的检测结果。然而，我们发现EBM训练过程中马尔可夫链蒙特卡洛（MCMC）采样的非混合性会削弱其检测性能。为解决这一问题，我们提出了一种基于能量校正的混合类别条件高斯分布模型。在CIFAR-10/CIFAR-100 OOD检测基准测试中，与KNN检测器等强基线方法相比，我们取得了更优的结果。