mlsanityData to 7 Oct 2026

Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theory

arXiv 2412.17152 ↗First posted 22 Dec 2024cs.LG stat.MLcited by 10

Not on the map. Why?

Abstract (arXiv)

Feature-based explanations, using perturbations or gradients, are a prevalent tool to understand decisions of black box machine learning models. Yet, differences between these methods still remain mostly unknown, which limits their applicability for practitioners. In this work, we introduce a unified framework for local and global feature-based explanations using two well-established concepts: functional ANOVA (fANOVA) from statistics, and the notion of value and interaction from cooperative game theory. We introduce three fANOVA decompositions that determine the influence of feature distributions, and use game-theoretic measures, such as the Shapley value and interactions, to specify the influence of higher-order interactions. Our framework combines these two dimensions to uncover similarities and differences between a wide range of explanation techniques for features and groups of features. We then empirically showcase the usefulness of our framework on synthetic and real-world datasets.

Key facts. Cited by 10 papers in the map (data through 7 Oct 2026). Builds on 43 of its 83 references that are in the map. First posted on arXiv 22 Dec 2024.

Builds on

43 of 83 references.

References · 43

  1. 2019PyTorch: An Imperative Style, High-Performance Deep Learning Librarycited by 20,160
  2. 2017A Unified Approach to Interpreting Model Predictionscited by 6,488
  3. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,120
  4. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,040
  5. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,243
  6. 2017Axiomatic Attribution for Deep Networkscited by 3,350
  7. 2013Visualizing and Understanding Convolutional Networkscited by 2,278
  8. 2017Learning Important Features Through Propagating Activation Differencescited by 1,454
  9. 2017SmoothGrad: removing noise by adding noisecited by 1,097
  10. 2017Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPRcited by 1,086
  11. 2018All Models are Wrong, but Many are Useful: Learning a Variable's Importance by Studying an Entire Class of Prediction Models Simultaneouslycited by 364
  12. 2019The many Shapley values for model explanationcited by 341
31 more
  1. 2018Consistent Individualized Feature Attribution for Tree Ensemblescited by 279
  2. 2020Explaining by Removing: A Unified Framework for Model Explanationcited by 218
  3. 2016Visualizing the Effects of Predictor Variables in Black Box Supervised Learning Modelscited by 212
  4. 2019Explaining individual predictions when features are dependent: More accurate approximations to Shapley valuescited by 205
  5. 2017Visualizing Deep Neural Network Decisions: Prediction Difference Analysiscited by 166
  6. 2019Feature relevance quantification in explainable AI: A causal problemcited by 164
  7. 2019The Shapley Taylor Interaction Indexcited by 143
  8. 2022Algorithms to estimate Shapley value feature attributionscited by 114
  9. 2020Explaining Explanations: Axiomatic Feature Interactions for Deep Networkscited by 103
  10. 2021Sampling Permutations for Shapley Value Estimationcited by 101
  11. 2022Which Explanation Should I Choose? A Function Approximation Perspective to Characterizing Post Hoc Explanationscited by 77
  12. 2020How does this interaction affect me? Interpretable attribution for feature interactionscited by 73
  13. 2022Faith-Shap: The Faithful Shapley Interaction Indexcited by 68
  14. 2022From Shapley Values to Generalized Additive Models and backcited by 59
  15. 2023SHAP-IQ: Unified Approximation of any-order Shapley Interactionscited by 50
  16. 2024shapiq: Shapley Interactions for Machine Learningcited by 42
  17. 2020Model-agnostic Feature Importance and Effects with Dependent Features -- A Conditional Subgroup Approachcited by 37
  18. 2024Beyond TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for Tree Ensemblescited by 30
  19. 2024KernelSHAP-IQ: Weighted Least-Square Optimization for Shapley Interactionscited by 22
  20. 2022Unifying local and global model explanations by functional decomposition of low dimensional structurescited by 18
  21. 2022REPID: Regional Effect Plots with implicit Interaction Detectioncited by 17
  22. 2021PredDiff: Explanations and Interactions from Conditional Expectationscited by 16
  23. 2024SVARM-IQ: Efficient Approximation of Any-order Shapley Interactions through Stratificationcited by 16
  24. 2021Joint Shapley values: a measure of joint feature importancecited by 15
  25. 2023A Comparative Study of Methods for Estimating Conditional Shapley Values and When to Use Themcited by 15
  26. 2021Grouped Feature Importance and Combined Features Effect Plotcited by 14
  27. 2024A Guide to Feature Importance Methods for Scientific Inferencecited by 11
  28. 2018Considerations When Learning Additive Explanations for Black-Box Modelscited by 10
  29. 2024Toward Understanding the Disagreement Problem in Neural Network Feature Attributioncited by 5
  30. 2024Efficient and Accurate Explanation Estimation with Distribution Compressioncited by 4
  31. 2024Interaction Decomposition of prediction functioncited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 4; 2026: 6.

10 papers.

  1. 2025Exact Shapley Attributions in Quadratic-time for FANOVA Gaussian Processescited by 6
  2. 2026PolySHAP: Extending KernelSHAP with Interaction-Informed Polynomial Regressioncited by 6
  3. 2025Amortized Linear-time Exact Shapley Value for Product-Kernel Methodscited by 3
  4. 2025What's Wrong with Your Synthetic Tabular Data? Using Explainable AI to Evaluate Generative Modelscited by 1
  5. 2025Permutation-Free High-Order Interaction Testscited by 1
  6. 2026Functional Decomposition and Shapley Interactions for Interpreting Survival Modelscited by 1
  7. 2026Attribution Upsampling should Redistribute, Not Interpolatecited by 0
  8. 2026Do covariates explain why these groups differ? The choice of reference group can reverse conclusions in the Oaxaca-Blinder decompositioncited by 0
  9. 2026A Hilbert-Valued Functional Decomposition Framework for Explaining Time-Dependent Outputscited by 0
  10. 2026A Hierarchy of Entropy-Shapley Games for Multivariate Predictive Uncertaintycited by 0