论文标题
通过学识渊博的关系现实的反事实解释
Realistic Counterfactual Explanations with Learned Relations
论文作者
论文摘要
许多现有的反事实解释方法忽略了数据属性之间的内在关系，因此无法产生现实的反事实。此外，现有的解释关系的模型需要域知识，这限制了其在复杂的现实世界应用中的适用性。在本文中，我们提出了一种新颖的方法来实现现实的反事实解释，以保留关系并最大程度地减少专家的干预措施。该模型直接通过具有最小域知识的变异自动编码器来了解关系，然后学会相应地扰动潜在空间。我们对合成数据集和现实数据集进行了广泛的实验。实验结果表明，所提出的模型从数据中学习关系，并将这些关系保存在生成的反事实中。特别是，它在Mahalanobis距离和约束可行性评分方面优于其他方法。
Many existing methods of counterfactual explanations ignore the intrinsic relationships between data attributes and thus fail to generate realistic counterfactuals. Moreover, the existing models that account for relationships require domain knowledge, which limits their applicability in complex real-world applications. In this paper, we propose a novel approach to realistic counterfactual explanations that preserve the relationships and minimise experts' interventions. The model directly learns the relationships by a variational auto-encoder with minimal domain knowledge and then learns to perturb the latent space accordingly. We conduct extensive experiments on both synthetic and real-world datasets. The experimental results demonstrate that the proposed model learns relationships from the data and preserves these relationships in generated counterfactuals. In particular, it outperforms other methods in terms of Mahalanobis distance, and the constraint feasibility score.
