Distributional Feature Importance for Correlated Predictors

2026-10-02
Normal 0 false false false TR X-NONE X-NONE /* Style Definitions */ table.MsoNormalTable {mso-style-name:"Normal Tablo"; mso-tstyle-rowband-size:0; mso-tstyle-colband-size:0; mso-style-noshow:yes; mso-style-priority:99; mso-style-parent:""; mso-padding-alt:0cm 5.4pt 0cm 5.4pt; mso-para-margin-top:0cm; mso-para-margin-right:0cm; mso-para-margin-bottom:8.0pt; mso-para-margin-left:0cm; line-height:115%; mso-pagination:widow-orphan; font-size:12.0pt; font-family:"Aptos",sans-serif; mso-ascii-font-family:Aptos; mso-ascii-theme-font:minor-latin; mso-hansi-font-family:Aptos; mso-hansi-theme-font:minor-latin; mso-font-kerning:1.0pt; mso-ligatures:standardcontextual; mso-ansi-language:TR;} In the literature, there exists a rapidly increasing trend in the use of machine learning and deep learning models thanks to their high predictive power and their ability to handle non-linearity and complex structures. However, because of their high number of dimensions, the inner workings of the models are obscured, making the model decisions mysterious. Model explanation can take several forms such as rule-logic extraction and counterfactual explanations, however explainability through feature importance is the most widely adopted strategy. SHAP and impurity-based importance are examples of the primary eXplainable AI (XAI) methods targeting feature importance. However, they suffer from the high level of relation between the variables, which is called multicollinearity, and these methods distribute marginal contributions among correlated variables. Hence, in this study, we explore a framework that focuses on feature importance under multicollinearity while mainly drawing on optimal transport theory, specifically the Wasserstein distance together with structural information among the variables. Rather than relying on marginal effects alone, the framework quantifies the feature importance based on distributional shift in the model’s behavior. Preliminary findings suggest that this integration is a promising alternative for explainability and interpretability, especially in the presence of multicollinearity.
22nd International Conference Applied Statistics 2026
Citation Formats
B. Koca and C. Yozgatlıgil, “Distributional Feature Importance for Correlated Predictors,” presented at the 22nd International Conference Applied Statistics 2026, Koper, Slovenya, 2026, Accessed: 00, 2026. [Online]. Available: https://as.mf.uni-lj.si/as2026book.pdf.