跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.21159 · v1

When Feature Representation Changes Local Explanations: A Controlled Comparison of SHAP and LIME under Numerical Discretization

首次提交 2026-11-16 22:23
版本更新 2026-11-16 22:23
雷达观测 2026-11-17 17:11
arXiv ↗PDF ↗

摘要

Local feature attributions are computed over a feature representation chosen by an upstream data-engineering step, yet they are read as statements about the underlying quantity. We ask how discretizing one continuous feature changes local KernelSHAP and LIME explanations when the classifier is held fixed, and whether the two explainers respond differently. A random forest is trained on UCI Adult and frozen; age is collapsed onto equi-width or quantile bin representatives (k in {3,5,10}) before the model receives it, and the same 120 held-out instances are explained in every condition across multiple seeds. We show analytically and numerically that for KernelSHAP this is exactly equivalent to explaining the unchanged model at the quantized instance against a quantized background, so the comparison is between two effective inputs rather than a pure explainer effect; for LIME no such equivalence holds. Each attribution change is compared with a seed-noise floor from re-running the explainer on the unchanged condition. KernelSHAP's age attribution exceeds its noise floor in all 6 pre-specified conditions (median ratios 2.4-59, all significant after Holm correction); on the raw scale LIME's exceeds its floor in only 2 of 6 and in the rest is no larger than its own run-to-run variation. SHAP's normalized change exceeds LIME's in every condition (rank-biserial 0.81-1.00), while LIME is 2.5-4.6x less consistent across repeats. Predicted classes change for only 0.8-5.8% of instances and held-out ROC-AUC never falls below 0.9097, yet 45-94% of instances show a large SHAP attribution change with no class flip. Feature-importance rankings barely move (median Spearman >= 0.979, top-3 overlap 1.00). The representation change and the change in effective input cannot be separated within a frozen-model design.

作者备注:20 pages, 7 figures, 6 tables. Code: https://github.com/Parvptl/shap-lime-representation-sensitivity

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-16 22:23 · 最新观测版本

When Feature Representation Changes Local Explanations: A Controlled Comparison of SHAP and LIME under Numerical Discretization

Local feature attributions are computed over a feature representation chosen by an upstream data-engineering step, yet they are read as statements about the underlying quantity. We ask how discretizing one continuous feature changes local KernelSHAP and LIME explanations when the classifier is held fixed, and whether the two explainers respond differently. A random forest is trained on UCI Adult and frozen; age is collapsed onto equi-width or quantile bin representatives (k in {3,5,10}) before the model receives it, and the same 120 held-out instances are explained in every condition across multiple seeds. We show analytically and numerically that for KernelSHAP this is exactly equivalent to explaining the unchanged model at the quantized instance against a quantized background, so the comparison is between two effective inputs rather than a pure explainer effect; for LIME no such equivalence holds. Each attribution change is compared with a seed-noise floor from re-running the explainer on the unchanged condition. KernelSHAP's age attribution exceeds its noise floor in all 6 pre-specified conditions (median ratios 2.4-59, all significant after Holm correction); on the raw scale LIME's exceeds its floor in only 2 of 6 and in the rest is no larger than its own run-to-run variation. SHAP's normalized change exceeds LIME's in every condition (rank-biserial 0.81-1.00), while LIME is 2.5-4.6x less consistent across repeats. Predicted classes change for only 0.8-5.8% of instances and held-out ROC-AUC never falls below 0.9097, yet 45-94% of instances show a large SHAP attribution change with no class flip. Feature-importance rankings barely move (median Spearman >= 0.979, top-3 overlap 1.00). The representation change and the change in effective input cannot be separated within a frozen-model design.

查看此版本解读 →查看此版本原文 ↗