We propose a method to estimate the uncertainty of the outcome of an image classifier on a given input datum. Deep neural networks commonly used for image classification are deterministic maps from an input image to an output class. As such, their outcome on a given datum involves no uncertainty, so we must specify what variability we are referring to when defining, measuring and interpreting uncertainty, and attributing "confidence" to the outcome. To this end, we introduce the Wellington Posterior, which is the distribution of outcomes that would have been obtained in response to data that could have been generated by the same scene that produced the given image. Since there are infinitely many scenes that could have generated any given image, the Wellington Posterior involves inductive transfer from scenes other than the one portrayed. We explore the use of data augmentation, dropout, ensembling, single-view reconstruction, and model linearization to compute a Wellington Posterior. Additional methods include the use of conditional generative models such as generative adversarial networks, neural radiance fields, and conditional prior networks. We test these methods against the empirical posterior obtained by performing inference on multiple images of the same underlying scene. These developments are only a small step towards assessing the reliability of deep network classifiers in a manner that is compatible with safety-critical applications and human interpretation.

翻译：我们提出了一种方法，用于估计图像分类器在给定输入数据上的结果不确定性。通常用于图像分类的深度神经网络是从输入图像到输出类别的确定性映射。因此，它们在给定数据上的结果不涉及任何不确定性，因此我们在定义、测量和解释不确定性，并将“置信度”归因于结果时，必须明确所指的变异性。为此，我们引入了惠灵顿后验，即对可能由产生给定图像的同一场景生成的数据进行响应时可能获得的结果分布。由于任何给定图像可能由无限多个场景生成，惠灵顿后验涉及对除所描绘场景外的其他场景的归纳迁移。我们探索了使用数据增强、丢弃法、集成学习、单视图重建和模型线性化来计算惠灵顿后验的方法。其他方法包括使用条件生成模型，如生成对抗网络、神经辐射场和条件先验网络。我们针对通过推断同一底层场景的多幅图像所获得的经验后验，对这些方法进行了测试。这些发展仅仅是在以与安全关键应用和人类解释兼容的方式评估深度网络分类器可靠性方面迈出的一小步。