arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2210.09933cs.LG
收藏

基于项目反应理论(eXirt)的解释:一种从信任视角解释树集成模型的模型特定方法

Explanations Based on Item Response Theory (eXirt): A Model-Specific Method to Explain Tree-Ensemble Model in Trust Perspective

Federal University of Pará (UFPA)(帕拉州联邦大学);Vale Institute of Technology (ITV)(淡水河谷技术研究院);Federal Institute of Education, Science and Technology of Pará (IFPA)(帕拉州联邦教育、科学和技术学院);University of Montpellier(蒙彼利埃大学);La Ligue Contre le Cancer(抗癌联盟)

José Ribeiro, Lucas Cardoso, Raíssa Silva, Vitor Cirilo, Níkolas Carneiro, Ronnie Alves

AI总结:

本文提出基于项目反应理论的新 XAI 方法 eXirt,面向二分类表格数据的树集成黑盒模型,通过全局特征排序和基于示例的局部解释提升模型可解释性与信任。

摘要:

近年来,可解释人工智能(XAI)研究者一直在形式化相关提案并开发新方法来解释黑盒模型,但社区对于应使用哪种方法解释这些模型尚未形成普遍共识,且这一选择几乎直接与特定方法的流行度相关。Ciu、Dalex、Eli5、Lofo、Shap 和 Skater 等方法相继出现,旨在通过特征相关性的全局排序来解释黑盒模型;这些方法基于不同方法论生成全局解释,以表明模型输入如何解释其预测。在此背景下,本研究使用 41 个数据集、4 种树集成算法(Light Gradient Boosting、CatBoost、Random Forest 和 Gradient Boosting)以及 6 种 XAI 方法,支持推出一种新的 XAI 方法——eXirt。该方法基于项目反应理论(IRT),面向使用表格数据并涉及二分类问题的树集成黑盒模型。在第一组分析中,将 eXirt 的 164 个全局特征相关性排序与文献中其他 XAI 方法的 984 个排序进行比较,以揭示它们之间的相似性与差异。在第二项分析中,展示了 eXirt 独有的基于示例解释(Explanation-by-example)的解释,有助于理解模型信任。由此验证,eXirt 能够生成树集成模型的全局解释,并通过 IRT 生成模型实例的局部解释,表明这一成熟理论可用于机器学习,以获得可解释且可靠的模型。

英文摘要:

In recent years, XAI researchers have been formalizing proposals and developing new methods to explain black box models, with no general consensus in the community on which method to use to explain these models, with this choice being almost directly linked to the popularity of a specific method. Methods such as Ciu, Dalex, Eli5, Lofo, Shap and Skater emerged with the proposal to explain black box models through global rankings of feature relevance, which based on different methodologies, generate global explanations that indicate how the model's inputs explain its predictions. In this context, 41 datasets, 4 tree-ensemble algorithms (Light Gradient Boosting, CatBoost, Random Forest, and Gradient Boosting), and 6 XAI methods were used to support the launch of a new XAI method, called eXirt, based on Item Response Theory - IRT and aimed at tree-ensemble black box models that use tabular data referring to binary classification problems. In the first set of analyses, the 164 global feature relevance ranks of the eXirt were compared with 984 ranks of the other XAI methods present in the literature, seeking to highlight their similarities and differences. In a second analysis, exclusive explanations of the eXirt based on Explanation-by-example were presented that help in understanding the model trust. Thus, it was verified that eXirt is able to generate global explanations of tree-ensemble models and also local explanations of instances of models through IRT, showing how this consolidated theory can be used in machine learning in order to obtain explainable and reliable models.

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。

↑