mlsanityData to 14 Oct 2026

Beyond TreeSHAP: Efficient Computation of Any-Order Shapley Interactions for Tree Ensembles

arXiv 2401.12069 ↗First posted 22 Jan 2024cs.LGcited by 31

Not on the map. Why?

Abstract (arXiv)

While shallow decision trees may be interpretable, larger ensemble models like gradient-boosted trees, which often set the state of the art in machine learning problems involving tabular data, still remain black box models. As a remedy, the Shapley value (SV) is a well-known concept in explainable artificial intelligence (XAI) research for quantifying additive feature attributions of predictions. The model-specific TreeSHAP methodology solves the exponential complexity for retrieving exact SVs from tree-based models. Expanding beyond individual feature attribution, Shapley interactions reveal the impact of intricate feature interactions of any order. In this work, we present TreeSHAP-IQ, an efficient method to compute any-order additive Shapley interactions for predictions of tree-based models. TreeSHAP-IQ is supported by a mathematical framework that exploits polynomial arithmetic to compute the interaction scores in a single recursive traversal of the tree, akin to Linear TreeSHAP. We apply TreeSHAP-IQ on state-of-the-art tree ensembles and explore interactions on well-established benchmark datasets.

Key facts. Cited by 31 papers in the map (data through 14 Oct 2026). Builds on 19 of its 50 references that are in the map. First posted on arXiv 22 Jan 2024.

Builds on

19 of 50 references.

References · 19

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2016XGBoost: A Scalable Tree Boosting Systemcited by 5,166
  3. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 443
  4. 2019The many Shapley values for model explanationcited by 344
  5. 2020Problems with Shapley-value-based explanations as feature importance measurescited by 200
  6. 2020Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regressioncited by 154
  7. 2019The Shapley Taylor Interaction Indexcited by 143
  8. 2020True to the Model or True to the Data?cited by 126
  9. 2021FastSHAP: Real-Time Shapley Value Estimationcited by 125
  10. 2022Algorithms to estimate Shapley value feature attributionscited by 115
  11. 2020How does this interaction affect me? Interpretable attribution for feature interactionscited by 73
  12. 2022Faith-Shap: The Faithful Shapley Interaction Indexcited by 69
7 more
  1. 2019OpenML-Python: an extensible Python API for OpenMLcited by 63
  2. 2022From Shapley Values to Generalized Additive Models and backcited by 60
  3. 2018Visualizing the Feature Importance for Black Box Modelscited by 55
  4. 2023SHAP-IQ: Unified Approximation of any-order Shapley Interactionscited by 51
  5. 2019Purifying Interaction Effects with the Functional ANOVA: An Efficient Algorithm for Recovering Identifiable Additive Modelscited by 42
  6. 2021Fast TreeSHAP: Accelerating SHAP Value Computation for Treescited by 29
  7. 2020High Dimensional Model Explanations: an Axiomatic Approachcited by 13

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 8; 2025: 11; 2026: 12.

31 papers.

Sort:

By area:

  1. 2024shapiq: Shapley Interactions for Machine Learningcited by 43
  2. 2024KernelSHAP-IQ: Weighted Least-Square Optimization for Shapley Interactionscited by 22
  3. 2025Exact Computation of Any-Order Shapley Interactions for Graph Neural Networkscited by 13
  4. 2024Unifying Feature-Based Explanations with Functional ANOVA and Cooperative Game Theorycited by 10
  5. 2025SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelismcited by 8
  6. 2025Tree Ensemble Explainability through the Hoeffding Functional Decomposition and TreeHFD Algorithmcited by 6
  7. 2026An Odd Estimator for Shapley Valuescited by 5
  8. 2024Efficient and Accurate Explanation Estimation with Distribution Compressioncited by 4
  9. 2024Fast Estimation of Partial Dependence Functions using Treescited by 4
  10. 2024Disentangling Interactions and Dependencies in Feature Attributioncited by 4
  11. 2025HyperSHAP: Shapley Values and Interactions for Explaining Hyperparameter Optimizationcited by 4
  12. 2025UbiQTree: Uncertainty Quantification in XAI with Tree Ensemblescited by 4
19 more
  1. 2026TreeGrad-Ranker: Feature Ranking via O(L)-Time Gradients for Decision Treescited by 4
  2. 2026Exactly Computing do-Shapley Valuescited by 3
  3. 2024A Comprehensive Study of Shapley Value in Data Analyticscited by 2
  4. 2025Adaptive Prompting: Ad-hoc Prompt Composition for Social Bias Detectioncited by 2
  5. 2026Quadrature-TreeSHAP: Depth-Independent TreeSHAP and Shapley Interactionscited by 2
  6. 2024Energy-Based Model for Accurate Estimation of Shapley Values in Feature Attributioncited by 1
  7. 2025Surrogate Interpretable Graph for Random Decision Forestscited by 1
  8. 2026Adalina: Adaptive Linear Approximation for the Shapley Value and Beyondcited by 1
  9. 2026Generalized Functional ANOVA: A Complete Theoretical Frameworkcited by 1
  10. 2025A novel gradient-based method for decision trees optimizing arbitrary differential loss functionscited by 0
  11. 2025CUBE: Contrastive Understanding by Balanced Experimentscited by 0
  12. 2025SX-GeoTree: Self-eXplaining Geospatial Regression Tree Incorporating the Spatial Similarity of Feature Attributionscited by 0
  13. 2025Provable Recovery of Locally Important Signed Features and Interactions from Random Forestcited by 0
  14. 2026Rethinking XAI Evaluation: A Human-Centered Audit of Shapley Benchmarks in High-Stakes Settingscited by 0
  15. 2026ConfoundingSHAP: Quantifying confounding strength in causal inferencecited by 0
  16. 2026TN-SHAP-G: Graph-Structured Tensor Network Surrogates for Shapley Values and Interactionscited by 0
  17. 2026OperatorSHAP: Fast and Accurate Shapley Value Estimation for Neural Operatorscited by 0
  18. 2026Nonlinear Axiomatic Attribution for Cooperative Gamescited by 0
  19. 2026Explaining Heterogeneous Treatment Effects with Causal Forests and the Hoeffding Functional Decompositioncited by 0