arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2209.05690cs.LGcs.CV
收藏

基于概念的表格数据解释

Concept-Based Explanations for Tabular Data

Varsha Pendyala, Jihye Choi

摘要:

机器学习模型的可解释性一直是安全部署机器学习系统的重要研究领域。一种特殊的方法是将模型决策归因于人类可以理解的高级概念。然而,深度神经网络(DNN)的这种基于概念的可解释性主要是在图像域上研究的。在本文中,我们扩展TCAV,概念属性的方法,表格学习,提供了一个想法,如何定义概念的表格数据。在一个具有地面真理概念解释的合成数据集和一个真实世界的数据集上,我们展示了我们的方法在生成与人类直觉相匹配的可解释性结果方面的有效性。最重要的是,我们提出了一个基于TCAV的公平性概念,它量化了DNN的哪一层已经学习了导致模型预测偏差的表示。此外,我们实证证明TCAV为基础的公平性的一组公平的概念,人口均等的关系。

英文摘要:

The interpretability of machine learning models has been an essential area of research for the safe deployment of machine learning systems. One particular approach is to attribute model decisions to high-level concepts that humans can understand. However, such concept-based explainability for Deep Neural Networks (DNNs) has been studied mostly on image domain. In this paper, we extend TCAV, the concept attribution approach, to tabular learning, by providing an idea on how to define concepts over tabular data. On a synthetic dataset with ground-truth concept explanations and a real-world dataset, we show the validity of our method in generating interpretability results that match the human-level intuitions. On top of this, we propose a notion of fairness based on TCAV that quantifies what layer of DNN has learned representations that lead to biased predictions of the model. Also, we empirically demonstrate the relation of TCAV-based fairness to a group fairness notion, Demographic Parity.

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。

↑