When Interpretability Is Unequally Distributed: Fairness in Hybrid Interpretable Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918528049741824 |
|---|---|
| author | Zare, Ziba Jabbar Aïvodji, Ulrich Ferry, Julien Vidal, Thibaut |
| author_facet | Zare, Ziba Jabbar Aïvodji, Ulrich Ferry, Julien Vidal, Thibaut |
| contents | Hybrid interpretable models combine a transparent component with a black-box model by assigning some examples to the former and deferring the rest to the latter. While this design enables flexible tradeoffs between accuracy and interpretability, it also raises a distinct procedural fairness concern: some demographic groups may systematically receive interpretable decisions, while others are disproportionately routed to a black box.
We formalize this issue as Interpretability Coverage Disparity (ICD), a demographic-parity-style measure applied to the routing decision of hybrid interpretable models. Using tools from predictive multiplicity, we study ICD across four hybrid interpretable learning methods, three standard fairness benchmark datasets, and multiple sensitive attributes. Our experiments reveal substantial ICD in intermediate transparency regimes, where both the interpretable and black-box components are actively used. We further show that simple coverage-disparity constraints can significantly reduce ICD in exact hybrid learning methods, with marginal impact on accuracy and sparsity. In several settings, ICD mitigation also improves standard algorithmic fairness metrics. These results show that hybrid interpretable models should be audited not only for predictive fairness, but also for how they allocate interpretability across individuals and groups. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_28626 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | When Interpretability Is Unequally Distributed: Fairness in Hybrid Interpretable Models Zare, Ziba Jabbar Aïvodji, Ulrich Ferry, Julien Vidal, Thibaut Machine Learning Hybrid interpretable models combine a transparent component with a black-box model by assigning some examples to the former and deferring the rest to the latter. While this design enables flexible tradeoffs between accuracy and interpretability, it also raises a distinct procedural fairness concern: some demographic groups may systematically receive interpretable decisions, while others are disproportionately routed to a black box. We formalize this issue as Interpretability Coverage Disparity (ICD), a demographic-parity-style measure applied to the routing decision of hybrid interpretable models. Using tools from predictive multiplicity, we study ICD across four hybrid interpretable learning methods, three standard fairness benchmark datasets, and multiple sensitive attributes. Our experiments reveal substantial ICD in intermediate transparency regimes, where both the interpretable and black-box components are actively used. We further show that simple coverage-disparity constraints can significantly reduce ICD in exact hybrid learning methods, with marginal impact on accuracy and sparsity. In several settings, ICD mitigation also improves standard algorithmic fairness metrics. These results show that hybrid interpretable models should be audited not only for predictive fairness, but also for how they allocate interpretability across individuals and groups. |
| title | When Interpretability Is Unequally Distributed: Fairness in Hybrid Interpretable Models |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2605.28626 |