Provable Recovery of Locally Important Signed Features and Interactions from Random Forest
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912759673782272 |
|---|---|
| author | Vuk, Kata Ihlo, Nicolas Alexander Behr, Merle |
| author_facet | Vuk, Kata Ihlo, Nicolas Alexander Behr, Merle |
| contents | Feature and Interaction Importance (FII) methods are essential in supervised learning for assessing the relevance of input variables and their interactions in complex prediction models. In many domains, such as personalized medicine, local interpretations for individual predictions are often required, rather than global scores summarizing overall feature importance. Random Forests (RFs) are widely used in these settings, and existing interpretability methods typically exploit tree structures and split statistics to provide model-specific insights. However, theoretical understanding of local FII methods for RF remains limited, making it unclear how to interpret high importance scores for individual predictions. We propose a novel, local, model-specific FII method that identifies frequent co-occurrences of features along decision paths, combining global patterns with those observed on paths specific to a given test point. We prove that our method consistently recovers the true local signal features and their interactions under a Locally Spike Sparse (LSS) model and also identifies whether large or small feature values drive a prediction. We illustrate the usefulness of our method and theoretical results through simulation studies and a real-world data example. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_11081 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Provable Recovery of Locally Important Signed Features and Interactions from Random Forest Vuk, Kata Ihlo, Nicolas Alexander Behr, Merle Machine Learning Methodology Feature and Interaction Importance (FII) methods are essential in supervised learning for assessing the relevance of input variables and their interactions in complex prediction models. In many domains, such as personalized medicine, local interpretations for individual predictions are often required, rather than global scores summarizing overall feature importance. Random Forests (RFs) are widely used in these settings, and existing interpretability methods typically exploit tree structures and split statistics to provide model-specific insights. However, theoretical understanding of local FII methods for RF remains limited, making it unclear how to interpret high importance scores for individual predictions. We propose a novel, local, model-specific FII method that identifies frequent co-occurrences of features along decision paths, combining global patterns with those observed on paths specific to a given test point. We prove that our method consistently recovers the true local signal features and their interactions under a Locally Spike Sparse (LSS) model and also identifies whether large or small feature values drive a prediction. We illustrate the usefulness of our method and theoretical results through simulation studies and a real-world data example. |
| title | Provable Recovery of Locally Important Signed Features and Interactions from Random Forest |
| topic | Machine Learning Methodology |
| url | https://arxiv.org/abs/2512.11081 |