Towards Rigorous Explainability by Feature Attribution 事件

Towards Rigorous Explainability by Feature Attribution arXiv:2604.15898v2 Announce Type: replace Abstract: For around a decade, non-symbolic methods have been the option of choice when explaining complex machine learning (ML) models. Unfortunately, such methods lack rigor and can mislead human decision-makers. In high-stakes uses of ML, the lack of rigor is especially problematic. One prime example of provable lack of rigor is the adoption of Shapley values in explainable artificial intelligenc