基于反事实分析的监督式特征压缩
Supervised Feature Compression based on Counterfactual Analysis
AI总结：
本文利用反事实解释检测黑盒模型的重要决策边界，对特征进行可调粒度的监督式离散化，进而训练出近似黑盒模型且可解释、紧凑的最优决策树，并在真实数据集上验证了其在准确性与稀疏性方面的有效性。
摘要：
反事实解释正成为事后可解释机器学习中的事实标准。对于给定的分类器和一个被分到不期望类别的实例，其反事实解释对应于对该实例的微小扰动，从而改变分类结果。本研究旨在利用反事实解释来检测预训练黑盒模型的重要决策边界。这些信息被用于构建数据集中特征的监督式离散化，并具有可调节的粒度。利用离散化后的数据集，可以训练一个最优决策树，该决策树既近似黑盒模型，又具有可解释性和紧凑性。在真实世界数据集上的数值结果表明，该方法在准确性和稀疏性方面具有有效性。
英文摘要：
Counterfactual Explanations are becoming a de-facto standard in post-hoc interpretable machine learning. For a given classifier and an instance classified in an undesired class, its counterfactual explanation corresponds to small perturbations of that instance that allows changing the classification outcome. This work aims to leverage Counterfactual Explanations to detect the important decision boundaries of a pre-trained black-box model. This information is used to build a supervised discretization of the features in the dataset with a tunable granularity. Using the discretized dataset, an optimal Decision Tree can be trained that resembles the black-box model, but that is interpretable and compact. Numerical results on real-world datasets show the effectiveness of the approach in terms of accuracy and sparsity.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
