arXiv：cs.LG（机器学习，全量分类）· Poushali Sengupta, Sabita Maharjan, Frank Eliassen, Shashi Raj Pandey, Yan Zhang·· 14 小时前AI 评分30
MCIR：一种具备可靠性保证的特征依赖感知可解释性方法
MCIR: A Feature Dependence-Aware Explainability Method with Reliability Guarantees
AI 导读
研究者提出依赖感知的全局特征重要性方法 MCIR-M，通过互相关影响比率（MCIR）将每个特征在强依赖邻居条件下归一化，得分落在 [0,1] 区间，在精确条件冗余时为零。该方法在合成冗余实验和 UCI HAR 基准上展现出依赖感知的排序行为，在注入近重复预测变量时优势最明显。论文已被 TMLR 接收。
正文
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 global feature-importance approach that quantifies the unique predictive information contributed by each feature beyond a selected dependence neighbourhood. MCIR-M introduces the Mutual Correlation Impact Ratio (MCIR), which conditions each feature on strongly dependent neighbours and computes a normalized ratio of conditional to block-level information. The population score lies in [0,1] and equals zero under exact conditional redundancy. We also introduce a lightweight estimation procedure that computes MCIR using a fraction of the available data and evaluates agreement with full-data explanations. Across controlled synthetic redundancy experiments and the UCI HAR benchmark, MCIR shows dependence-aware ranking behaviour, with its clearest advantage under injected near-duplicate predictors. Comparisons with independent and conditional SHAP, SAGE, HSIC, MI-based scores, and CIR-family baselines are mixed across real-data criteria. Reduced explanation samples lower computational burden in the evaluated configurations, while agreement with full-data explanations is assessed separately through ranking, head-set, and faithfulness diagnostics. Overall, MCIR-M provides a practical dependence-aware diagnostic for global explanation under strong feature dependence.
| Comments: | Accepted for publication in Transactions on Machine Learning Research (TMLR) |
| Subjects: | Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computation (stat.CO) |
| Cite as: | arXiv:2610.01641 [cs.LG] |
| (or arXiv:2610.01641v1 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2610.01641 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Poushali Sengupta [view email]
[v1]
Thu, 1 Oct 2026 13:07:01 UTC (9,739 KB)
来源：arXiv：cs.LG（机器学习，全量分类） · arxiv.org