MCIR: A Feature Dependence-Aware Explainability Method with Reliability Guarantees

Excerpt

arXiv:2610.01641v1 Announce Type: new Abstract: Modern machine-learning models often contain strongly dependent or redundant features, making feature attribution difficult because shared predictive information can be distributed across correlated predictors. Existing methods such as SHAP, LIME, HSIC, MI/CMI, and SAGE may therefore produce unstable rankings under multicollinearity or near-duplicate predictors. We propose the Mutual Correlation Impact Ratio Method (MCIR-M), a dependence-aware glob