论文标题
获得线索：一种解释不确定性估计的方法
Getting a CLUE: A Method for Explaining Uncertainty Estimates
论文作者
论文摘要
不确定性估计和解释性都是值得信赖的机器学习系统的重要因素。但是，这两个领域的交集几乎没有工作。我们通过提出一种新方法来解释从贝叶斯神经网络（BNNS）等可区分概率模型中解释不确定性估计的新方法。我们的方法，反事实的潜在不确定性解释（线索）指示如何更改输入，同时将其保留在数据歧管上，以使BNN对输入的预测更加自信。我们通过1）验证线索，一个新的框架，用于评估不确定性的反事实解释，2）一系列消融实验，以及3）用户研究。我们的实验表明，线索的表现优于基准，并且使从业者能够更好地了解哪些输入模式负责预测不确定性。
Both uncertainty estimation and interpretability are important factors for trustworthy machine learning systems. However, there is little work at the intersection of these two areas. We address this gap by proposing a novel method for interpreting uncertainty estimates from differentiable probabilistic models, like Bayesian Neural Networks (BNNs). Our method, Counterfactual Latent Uncertainty Explanations (CLUE), indicates how to change an input, while keeping it on the data manifold, such that a BNN becomes more confident about the input's prediction. We validate CLUE through 1) a novel framework for evaluating counterfactual explanations of uncertainty, 2) a series of ablation experiments, and 3) a user study. Our experiments show that CLUE outperforms baselines and enables practitioners to better understand which input patterns are responsible for predictive uncertainty.
