论文标题

不确定性量化中标记了什么?不确定性分类的潜在密度模型

What is Flagged in Uncertainty Quantification? Latent Density Models for Uncertainty Categorization

论文作者

Sun, Hao, van Breugel, Boris, Crabbe, Jonathan, Seedat, Nabeel, van der Schaar, Mihaela

论文摘要

不确定性量化(UQ)对于创建值得信赖的机器学习模型至关重要。近年来,UQ方法急剧上升,可以标记可疑示例,但是,通常不清楚这些方法确切地识别出什么。在这项工作中,我们提出了一个框架,用于对UQ方法在分类任务中标记的不确定示例进行分类。我们介绍了混淆密度矩阵 - 基于内核的错误分类密度的近似 - 并将其用于将通过给定不确定方法方法确定的可疑示例分为三个类:分布(OOD)示例过量(BND)示例(BND)示例,以及在分布分布错误分类错误(IDM)中的示例。通过广泛的实验,我们表明我们的框架为评估不确定性量化方法之间的差异提供了一种新的和不同的观点,从而形成了宝贵的评估基准。

Uncertainty Quantification (UQ) is essential for creating trustworthy machine learning models. Recent years have seen a steep rise in UQ methods that can flag suspicious examples, however, it is often unclear what exactly these methods identify. In this work, we propose a framework for categorizing uncertain examples flagged by UQ methods in classification tasks. We introduce the confusion density matrix -- a kernel-based approximation of the misclassification density -- and use this to categorize suspicious examples identified by a given uncertainty method into three classes: out-of-distribution (OOD) examples, boundary (Bnd) examples, and examples in regions of high in-distribution misclassification (IDM). Through extensive experiments, we show that our framework provides a new and distinct perspective for assessing differences between uncertainty quantification methods, thereby forming a valuable assessment benchmark.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源