Explanation Multiplicity in SHAP: Characterization and Assessment
Quick summary
arXiv:2601.12654v3 Announce Type: replace-cross Abstract: SHAP explanations are widely used in high-stakes settings to justify decisions, yet they can differ substantially across repeated runs, even when the model, the input instance, and the prediction are held fixed. Prior work has documented disagreement between explanation methods; we show that substantial disagreement arises even within SHAP across reruns of the same estimator on the same trained model and instance. We call this phenomenon explanation multiplicity and develop an evaluation methodology for characterizing it under deploymen
Key takeaways
- arXiv:2601.12654v3 Announce Type: replace-cross Abstract: SHAP explanations are widely used in high-stakes settings to justify decisions, yet they can differ substantially across repeated runs, even when the model, the input instance, and the prediction are held fixed.
- Prior work has documented disagreement between explanation methods; we show that substantial disagreement arises even within SHAP across reruns of the same estimator on the same trained model and instance.
- We call this phenomenon explanation multiplicity and develop an evaluation methodology for characterizing it under deploymen
Why it matters
“Explanation Multiplicity in SHAP: Characterization and Assessment” highlights the need for repeatable measurement rather than a single impressive demonstration. Independent validation across datasets and clearly stated limitations determine whether a result can guide product decisions.
Member comments