English

Trustworthy Feature Importance Avoids Unrestricted Permutations

Machine Learning 2026-04-14 v1 Machine Learning

Abstract

Feature importance methods using unrestricted permutations are flawed due to extrapolation errors; such errors appear in all non-trivial variable importance approaches. We propose three new approaches: conditional model reliance and Knockoffs with Gaussian transformation, and restricted ALE plot designs. Theoretical and numerical results show our strategies reduce/eliminate extrapolation.

Keywords

Cite

@article{arxiv.2604.11253,
  title  = {Trustworthy Feature Importance Avoids Unrestricted Permutations},
  author = {Emanuele Borgonovo and Francesco Cappelli and Xuefei Lu and Elmar Plischke and Cynthia Rudin},
  journal= {arXiv preprint arXiv:2604.11253},
  year   = {2026}
}
R2 v1 2026-07-01T12:06:01.275Z