Conditional Feature Importance revisited: Double Robustness, Efficiency and Inference

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Reyero-Lobo, Angel, Neuvial, Pierre, Thirion, Bertrand
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914296662851584
author Reyero-Lobo, Angel
Neuvial, Pierre
Thirion, Bertrand
author_facet Reyero-Lobo, Angel
Neuvial, Pierre
Thirion, Bertrand
contents Conditional Feature Importance (CFI) is a classical variable importance measure that accounts for the relationship between the studied feature and the others. However, CFI has not yet been studied from a theoretical perspective because the conditional sampling step has generally been considered a purely practical problem. In this article, we demonstrate that the recent Conditional Permutation Importance (CPI) is indeed a valid implementation of this concept. Under the conditional null hypothesis, we then establish a double robustness property that can be used for variable selection. With either a valid model or a valid conditional sampler, the method correctly identifies null coordinates. Under the alternative hypothesis, we study the theoretical target and link it to the popular Total Sobol Index (TSI). We introduce the Sobol-CPI, which generalizes CPI/CFI, prove that it is nonparametrically efficient, and provide a bias correction. Finally, we propose a consistent and valid type-I error test and present numerical experiments that illustrate our findings.
format Preprint
id arxiv_https___arxiv_org_abs_2501_17520
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conditional Feature Importance revisited: Double Robustness, Efficiency and Inference
Reyero-Lobo, Angel
Neuvial, Pierre
Thirion, Bertrand
Statistics Theory
Conditional Feature Importance (CFI) is a classical variable importance measure that accounts for the relationship between the studied feature and the others. However, CFI has not yet been studied from a theoretical perspective because the conditional sampling step has generally been considered a purely practical problem. In this article, we demonstrate that the recent Conditional Permutation Importance (CPI) is indeed a valid implementation of this concept. Under the conditional null hypothesis, we then establish a double robustness property that can be used for variable selection. With either a valid model or a valid conditional sampler, the method correctly identifies null coordinates. Under the alternative hypothesis, we study the theoretical target and link it to the popular Total Sobol Index (TSI). We introduce the Sobol-CPI, which generalizes CPI/CFI, prove that it is nonparametrically efficient, and provide a bias correction. Finally, we propose a consistent and valid type-I error test and present numerical experiments that illustrate our findings.
title Conditional Feature Importance revisited: Double Robustness, Efficiency and Inference
topic Statistics Theory
url https://arxiv.org/abs/2501.17520