mlsanityData to 7 Oct 2026

Feature Importance Depends on Properties of the Data: Towards Choosing the Correct Explanations for Your Data and Decision Trees based Models

arXiv 2502.07153 ↗First posted 11 Feb 2025cs.LG cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

In order to ensure the reliability of the explanations of machine learning models, it is crucial to establish their advantages and limits and in which case each of these methods outperform. However, the current understanding of when and how each method of explanation can be used is insufficient. To fill this gap, we perform a comprehensive empirical evaluation by synthesizing multiple datasets with the desired properties. Our main objective is to assess the quality of feature importance estimates provided by local explanation methods, which are used to explain predictions made by decision tree-based models. By analyzing the results obtained from synthetic datasets as well as publicly available binary classification datasets, we observe notable disparities in the magnitude and sign of the feature importance estimates generated by these methods. Moreover, we find that these estimates are sensitive to specific properties present in the data. Although some model hyper-parameters do not significantly influence feature importance assignment, it is important to recognize that each method of explanation has limitations in specific contexts. Our assessment highlights these limitations and provides valuable insight into the suitability and reliability of different explanatory methods in various scenarios.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 22 of its 33 references that are in the map. First posted on arXiv 11 Feb 2025.

Builds on

22 of 33 references.

References · 22

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  2. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  3. 2018RISE: Randomized Input Sampling for Explanation of Black-box Modelscited by 825
  4. 2020Captum: A unified and generic model interpretability library for PyTorchcited by 542
  5. 2018On the Robustness of Interpretability Methodscited by 310
  6. 2018Consistent Individualized Feature Attribution for Tree Ensemblescited by 279
  7. 2022Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyondcited by 171
  8. 2022The Disagreement Problem in Explainable Machine Learning: A Practitioner's Perspectivecited by 156
  9. 2023The Clock and the Pizza: Two Stories in Mechanistic Explanation of Neural Networkscited by 153
  10. 2021Benchmarking and Survey of Explanation Methods for Black Box Modelscited by 136
  11. 2020Reliable Post hoc Explanations: Modeling Uncertainty in Explainabilitycited by 125
  12. 2020Benchmarking Deep Learning Interpretability in Time Series Predictionscited by 102
10 more
  1. 2019Benchmarking Attribution Methods with Relative Feature Importancecited by 89
  2. 2022Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanationscited by 77
  3. 2021Synthetic Benchmarks for Scientific Research in Explainable Machine Learningcited by 59
  4. 2022Evaluation of post-hoc interpretability methods in time-series classificationcited by 53
  5. 2019Can I Trust the Explainer? Verifying Post-hoc Explanatory Methodscited by 37
  6. 2021Order in the Court: Explainable AI Methods Prone to Disagreementcited by 36
  7. 2022ferret: a Framework for Benchmarking Explainers on Transformerscited by 26
  8. 2019A Debiased MDI Feature Importance Measure for Random Forestscited by 24
  9. 2022Comparing Explanation Methods for Traditional Machine Learning Models Part 1: An Overview of Current Methods and Quantifying Their Disagreementcited by 14
  10. 2022The impact of feature importance methods on the interpretation of defect classifierscited by 12

Built on by

Not cited yet.