论文标题
基于能量的分布检测
Energy-based Out-of-distribution Detection
论文作者
论文摘要
确定输入是否是分发（OOD）是在开放世界中安全部署机器学习模型的必不可少的基础。但是，以前依赖于SoftMax置信度得分的方法受到OOD数据的过度自信后分布。我们提出了使用能量评分的OOD检测统一框架。我们表明，与使用SoftMax分数的传统方法相比，能量得分更好地区分分布样本和分布样本。与SoftMax置信度分数不同，理论上能量评分与输入的概率密度一致，并且不容易受到过度自信问题的影响。在此框架内，可以灵活地将能量用作任何预训练的神经分类器以及可训练的成本函数的评分函数，以明确塑造能量表面以进行OOD检测。与SoftMax置信度得分相比，使用能量得分的CIFAR-10预先训练的宽带网将平均FPR（TPR 95％）降低18.03％。通过基于能量的培训，我们的方法的表现优于公共基准的最新方法。
Determining whether inputs are out-of-distribution (OOD) is an essential building block for safely deploying machine learning models in the open world. However, previous methods relying on the softmax confidence score suffer from overconfident posterior distributions for OOD data. We propose a unified framework for OOD detection that uses an energy score. We show that energy scores better distinguish in- and out-of-distribution samples than the traditional approach using the softmax scores. Unlike softmax confidence scores, energy scores are theoretically aligned with the probability density of the inputs and are less susceptible to the overconfidence issue. Within this framework, energy can be flexibly used as a scoring function for any pre-trained neural classifier as well as a trainable cost function to shape the energy surface explicitly for OOD detection. On a CIFAR-10 pre-trained WideResNet, using the energy score reduces the average FPR (at TPR 95%) by 18.03% compared to the softmax confidence score. With energy-based training, our method outperforms the state-of-the-art on common benchmarks.
