Researchers have introduced the $\phi$-table, a new method for statistically explaining global SHAP values in tabular black-box regression models. This approach moves beyond simple feature importance rankings to provide a more comprehensive understanding of model behavior. The $\phi$-table integrates SHAP importance with coefficients from a standardized linear surrogate, offering insights into the directionality of feature effects, their uncertainty, the fidelity of the surrogate model, and the stability of coefficients. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Introduces a novel statistical explanation method for SHAP values, enhancing interpretability of black-box models.
RANK_REASON This is a research paper introducing a new statistical explanation method for machine learning models. [lever_c_demoted from research: ic=1 ai=1.0]