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

TMLR Paper6812 Authors

Abstract: As modern machine learning models are deployed in high-stakes, data-rich environments, the interactions among features have grown more intricate and less amenable to traditional interpretation. Many explanation methods fail when features are strongly dependent. In the presence of multicollinearity or near-duplicate predictors, existing value attribution tools such as SHAP, LIME, HSIC, MI/CMI, and SAGE often distribute importance across redundant features, obscuring which variables represent "important and unique information". This may lead to unstable rankings, jeopardising importance scores, and usually results in a high computational cost. Recent correlation-aware approaches, such as CIR or BlockCIR, offer partial improvements but still struggle to fully separate redundancy from unique contributions at the feature level. To address this, we propose the Mutual Correlation Impact Ratio Method (MCIR-M), a simple and robust measure of global importance under feature dependence. MCIR-M introduces the score Mutual Correlation Impact Ratio (MCIR) that conditions each feature on a small set of its most correlated neighbours and computes a normalized ratio of conditional information having value range, \([0,1]\), which is comparable across tasks, and collapses to zero when a feature is redundant, enabling clear redundancy detection. In addition to MICR, we introduce a lightweight estimation procedure that requires only a fraction of the data while preserving the attribution behaviour of the full model. Across a synthetic household-energy dataset and the real UCI HAR benchmark, MCIR yields more stable and dependence-aware rankings than SHAP (independent and conditional), SAGE, HSIC, MI-based scores, and correlation-aware baselines such as CIR or BlockCIR. Lightweight explanations preserve over \(95\%\) top-feature agreement and reduce runtime by manyfold. These results demonstrate that MCIR-M provides a practical and scalable solution for global explanation in settings with strong feature dependence.

Submission Type: Long submission (more than 12 pages of main content)

Assigned Action Editor: ~Sivan_Sabato1

Submission Number: 6812