mlsanityData to 7 Oct 2026

A Guide to Feature Importance Methods for Scientific Inference

arXiv 2404.12862 ↗First posted 19 Apr 2024stat.ML cs.LG math.STcited by 11

Not on the map. Why?

Abstract (arXiv)

While machine learning (ML) models are increasingly used due to their high predictive power, their use in understanding the data-generating process (DGP) is limited. Understanding the DGP requires insights into feature-target associations, which many ML models cannot directly provide due to their opaque internal mechanisms. Feature importance (FI) methods provide useful insights into the DGP under certain conditions. Since the results of different FI methods have different interpretations, selecting the correct FI method for a concrete use case is crucial and still requires expert knowledge. This paper serves as a comprehensive guide to help understand the different interpretations of global FI methods. Through an extensive review of FI methods and providing new proofs regarding their interpretation, we facilitate a thorough understanding of these methods and formulate concrete recommendations for scientific inference. We conclude by discussing options for FI uncertainty estimation and point to directions for future research aiming at full statistical inference from black-box ML models.

Key facts. Cited by 11 papers in the map (data through 7 Oct 2026). Builds on 30 of its 62 references that are in the map. First posted on arXiv 19 Apr 2024.

Builds on

30 of 62 references.

References · 30

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  2. 2018A Survey Of Methods For Explaining Black Box Modelscited by 970
  3. 2016Distribution-Free Predictive Inference For Regressioncited by 657
  4. 2020AutoGluon-Tabular: Robust and Accurate AutoML for Structured Datacited by 368
  5. 2018All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneouslycited by 364
  6. 2018Consistent Individualized Feature Attribution for Tree Ensemblescited by 279
  7. 2020Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Surveycited by 252
  8. 2020Explaining by Removing: A Unified Framework for Model Explanationcited by 218
  9. 2016Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Modelscited by 212
  10. 2022The Disagreement Problem in Explainable Machine Learning: A Practitioner's Perspectivecited by 156
  11. 2022OpenXAI: Towards a Transparent Evaluation of Model Explanationscited by 125
  12. 2019Unrestricted Permutation forces Extrapolation: Variable Importance Requires at least One More Model, or There Is No Free Variable Importancecited by 85
18 more
  1. 2022Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanationscited by 77
  2. 2020General Pitfalls of Model-Agnostic Interpretation Methods for Machine Learning Modelscited by 74
  3. 2018A Simple and Effective Model-Based Variable Importance Measurecited by 50
  4. 2020A general framework for inference on algorithm-agnostic variable importancecited by 44
  5. 2020Model-agnostic Feature Importance and Effects with Dependent Features -- A Conditional Subgroup Approachcited by 37
  6. 2021How to avoid machine learning pitfalls: a guide for academic researcherscited by 34
  7. 2023The Rashomon Importance Distribution: Getting RID of Unstable, Single Model-based Variable Importancecited by 29
  8. 2019Testing Conditional Independence in Supervised Learning Algorithmscited by 27
  9. 2021Relating the Partial Dependence Plot and Permutation Feature Importance to the Data Generating Processcited by 24
  10. 2023Dear XAI Community, We Need to Talk! Fundamental Misconceptions in Current XAI Researchcited by 22
  11. 2019Variable Importance Clouds: A Way to Explore Variable Importance for the Set of Good Modelscited by 19
  12. 2022Unifying local and global model explanations by functional decomposition of low dimensional structurescited by 18
  13. 2023Decomposing Global Feature Effects Based on Feature Interactionscited by 17
  14. 2021Grouped Feature Importance and Combined Features Effect Plotcited by 14
  15. 2020Relative Feature Importancecited by 12
  16. 2021Shapley variable importance clouds for interpretable machine learningcited by 7
  17. 2022Conditional Feature Importance for Mixed Datacited by 7
  18. 2023Efficient SAGE Estimation via Causal Structure Learningcited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 5; 2026: 3.

11 papers.

  1. 2024Position: A Call to Action for a Human-Centered AutoML Paradigmcited by 14
  2. 2024Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theorycited by 10
  3. 2025A Statistical Case Against Empirical Human-AI Alignmentcited by 6
  4. 2025Deep Polynomial Chaos Expansioncited by 3
  5. 2024Which Imputation Fits Which Feature Selection Method? A Survey-Based Simulation Studycited by 2
  6. 2025Privilege Scorescited by 1
  7. 2025How to safely discard features based on aggregate SHAP valuescited by 1
  8. 2026xplainfi: Feature Importance and Statistical Inference for Machine Learning in Rcited by 1
  9. 2026CASHomon Sets: Efficient Rashomon Sets Across Multiple Model Classes and their Hyperparameterscited by 1
  10. 2025Feature-Guided Neighbor Selection for Non-Expert Evaluation of Model Predictionscited by 0
  11. 2026A Stationary-Distribution Theory for Triplet-Based Plateau Search in Random Forest Ensemble-Size Selectioncited by 0