Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2404.03506 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911827911245824 |
|---|---|
| author | Dandl, Susanne Blesch, Kristin Freiesleben, Timo König, Gunnar Kapar, Jan Bischl, Bernd Wright, Marvin |
| author_facet | Dandl, Susanne Blesch, Kristin Freiesleben, Timo König, Gunnar Kapar, Jan Bischl, Bernd Wright, Marvin |
| contents | Counterfactual explanations elucidate algorithmic decisions by pointing to scenarios that would have led to an alternative, desired outcome. Giving insight into the model's behavior, they hint users towards possible actions and give grounds for contesting decisions. As a crucial factor in achieving these goals, counterfactuals must be plausible, i.e., describing realistic alternative scenarios within the data manifold. This paper leverages a recently developed generative modeling technique -- adversarial random forests (ARFs) -- to efficiently generate plausible counterfactuals in a model-agnostic way. ARFs can serve as a plausibility measure or directly generate counterfactual explanations. Our ARF-based approach surpasses the limitations of existing methods that aim to generate plausible counterfactual explanations: It is easy to train and computationally highly efficient, handles continuous and categorical data naturally, and allows integrating additional desiderata such as sparsity in a straightforward manner. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2404_03506 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CountARFactuals -- Generating plausible model-agnostic counterfactual explanations with adversarial random forests Dandl, Susanne Blesch, Kristin Freiesleben, Timo König, Gunnar Kapar, Jan Bischl, Bernd Wright, Marvin Machine Learning Counterfactual explanations elucidate algorithmic decisions by pointing to scenarios that would have led to an alternative, desired outcome. Giving insight into the model's behavior, they hint users towards possible actions and give grounds for contesting decisions. As a crucial factor in achieving these goals, counterfactuals must be plausible, i.e., describing realistic alternative scenarios within the data manifold. This paper leverages a recently developed generative modeling technique -- adversarial random forests (ARFs) -- to efficiently generate plausible counterfactuals in a model-agnostic way. ARFs can serve as a plausibility measure or directly generate counterfactual explanations. Our ARF-based approach surpasses the limitations of existing methods that aim to generate plausible counterfactual explanations: It is easy to train and computationally highly efficient, handles continuous and categorical data naturally, and allows integrating additional desiderata such as sparsity in a straightforward manner. |
| title | CountARFactuals -- Generating plausible model-agnostic counterfactual explanations with adversarial random forests |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2404.03506 |