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 2406.15085cs.CL
收藏

通过统一诊断评估框架评估输入特征解释

Evaluating Input Feature Explanations through a Unified Diagnostic Evaluation Framework

University of Copenhagen(哥本哈根大学)

Jingyi Sun, Pepa Atanasova, Isabelle Augenstein

AI总结:

本文提出统一诊断评估框架,系统比较词元级、词元交互级和片段交互级三类输入特征解释,发现交互式片段解释在多数诊断属性上表现更优,并指出需进一步研究及融合不同解释类型以提升整体效果。

摘要:

解释机器学习模型的决策过程对于确保其可靠性和对最终用户的透明度至关重要。一种流行的解释形式突出关键输入特征,例如 i) 词元(如 Shapley Values 和 Integrated Gradients),ii) 词元之间的交互(如 Bivariate Shapley 和基于注意力的方法),或 iii) 输入片段之间的交互(如 Louvain Span Interactions)。然而,这些解释类型仅被孤立地研究,难以判断它们各自的适用性。为弥合这一差距,我们开发了一个统一框架,该框架有助于在包含四个诊断属性的基础上对突出解释和交互解释进行自动化和直接比较。我们在两种数据集和两种模型上对这三类输入特征解释——每类使用三种不同的解释技术——进行了广泛分析,并揭示出每种解释在不同诊断属性上具有不同的优势。尽管如此,交互式片段解释在大多数诊断属性上优于其他类型的输入特征解释。尽管相对研究不足,我们的分析强调需要进一步研究以改进生成这些解释类型的方法。此外,将它们在特定特征上表现更好的其他解释类型进行整合,可以进一步增强其整体有效性。

英文摘要:

Explaining the decision-making process of machine learning models is crucial for ensuring their reliability and transparency for end users. One popular explanation form highlights key input features, such as i) tokens (e.g., Shapley Values and Integrated Gradients), ii) interactions between tokens (e.g., Bivariate Shapley and Attention-based methods), or iii) interactions between spans of the input (e.g., Louvain Span Interactions). However, these explanation types have only been studied in isolation, making it difficult to judge their respective applicability. To bridge this gap, we develop a unified framework that facilitates an automated and direct comparison between highlight and interactive explanations comprised of four diagnostic properties. We conduct an extensive analysis across these three types of input feature explanations -- each utilizing three different explanation techniques -- across two datasets and two models, and reveal that each explanation has distinct strengths across the different diagnostic properties. Nevertheless, interactive span explanations outperform other types of input feature explanations across most diagnostic properties. Despite being relatively understudied, our analysis underscores the need for further research to improve methods generating these explanation types. Additionally, integrating them with other explanation types that perform better in certain characteristics could further enhance their overall effectiveness.

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。

↑