PredDiff is a model-agnostic, local attribution method that is firmly rooted in probability theory. Its simple intuition is to measure prediction changes while marginalizing features. In this work, we clarify properties of PredDiff and its close connection to Shapley values. We stress important differences between classification and regression, which require a specific treatment within both formalisms. We extend PredDiff by introducing a new, well-founded measure for interaction effects between arbitrary feature subsets. The study of interaction effects represents an inevitable step towards a comprehensive understanding of black-box models and is particularly important for science applications. Equipped with our novel interaction measure, PredDiff is a promising model-agnostic approach for obtaining reliable, numerically inexpensive and theoretically sound attributions.

PredDiff 是一种与模型无关的局部归因方法，它深深植根于概率论。它的简单直觉是在边缘化特征的同时测量预测变化。在这项工作中，我们阐明了 PredDiff 的属性及其与 Shapley 值的密切联系。我们强调分类和回归之间的重要区别，这需要在两种形式中进行特定处理。我们通过引入一种新的、有根据的测量任意特征子集之间的交互效应来扩展 PredDiff。交互效应的研究是全面理解黑盒模型的必然步骤，对于科学应用尤为重要。配备我们新颖的交互测量，PredDiff 是一种有前途的模型无关方法，用于获得可靠的、