论文标题
基于基于流的生成模型的基于kullback-leibler差异分布检测
Kullback-Leibler Divergence-Based Out-of-Distribution Detection with Flow-Based Generative Models
论文作者
论文摘要
最近的研究表明，与分布（ID）数据相比，包括基于流的模型和变异自动编码器在内的深层生成模型可能会将更高的可能性分配给分布（OOD）数据。但是，我们无法从模型中采样OOD数据。这种违反直觉现象尚未得到令人满意的解释，并通过基于流的模型给OOD检测带来了障碍。在本文中，我们证明定理研究基于流的模型中的kullback-leibler差异，并为上述现象提供了两个解释。基于我们的理论分析，我们提出了一种新方法\ padMethod \来利用表示形式的局部像素依赖性来执行异常检测。普遍基准的实验结果证明了我们方法的有效性和鲁棒性。对于组异常检测，我们的方法平均达到98.1 \％AUROC，小批量大小为5。相反，基于挑战性问题的基线典型测试方法平均仅实现64.6 \％AUROC。我们的方法还以9.1 \％AUROC优于最先进的方法。为了通过点异常检测，我们的方法平均达到90.7 \％的AUROC，并以5.2 \％的AUROC优于基线。此外，我们的方法具有最低值得注意的失败，并且是最强大的失败。
Recent research has revealed that deep generative models including flow-based models and Variational Autoencoders may assign higher likelihoods to out-of-distribution (OOD) data than in-distribution (ID) data. However, we cannot sample OOD data from the model. This counterintuitive phenomenon has not been satisfactorily explained and brings obstacles to OOD detection with flow-based models. In this paper, we prove theorems to investigate the Kullback-Leibler divergence in flow-based model and give two explanations for the above phenomenon. Based on our theoretical analysis, we propose a new method \PADmethod\ to leverage KL divergence and local pixel dependence of representations to perform anomaly detection. Experimental results on prevalent benchmarks demonstrate the effectiveness and robustness of our method. For group anomaly detection, our method achieves 98.1\% AUROC on average with a small batch size of 5. On the contrary, the baseline typicality test-based method only achieves 64.6\% AUROC on average due to its failure on challenging problems. Our method also outperforms the state-of-the-art method by 9.1\% AUROC. For point-wise anomaly detection, our method achieves 90.7\% AUROC on average and outperforms the baseline by 5.2\% AUROC. Besides, our method has the least notable failures and is the most robust one.
