mlsanityData to 14 Oct 2026

On Locality of Local Explanation Models

arXiv 2106.14648 ↗First posted 24 Jun 2021cs.LG stat.CO stat.MEcited by 25

Not on the map. Why?

Abstract (arXiv)

Shapley values provide model agnostic feature attributions for model outcome at a particular instance by simulating feature absence under a global population distribution. The use of a global population can lead to potentially misleading results when local model behaviour is of interest. Hence we consider the formulation of neighbourhood reference distributions that improve the local interpretability of Shapley values. By doing so, we find that the Nadaraya-Watson estimator, a well-studied kernel regressor, can be expressed as a self-normalised importance sampling estimator. Empirically, we observe that Neighbourhood Shapley values identify meaningful sparse feature relevance attributions that provide insight into local model behaviour, complimenting conventional Shapley analysis. They also increase on-manifold explainability and robustness to the construction of adversarial classifiers.

Key facts. Cited by 25 papers in the map (data through 14 Oct 2026). Builds on 27 of its 49 references that are in the map. First posted on arXiv 24 Jun 2021.

Builds on

27 of 49 references.

References · 27

  1. 2017A Unified Approach to Interpreting Model Predictionscited by 6,555
  2. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  3. 2017Axiomatic Attribution for Deep Networkscited by 3,384
  4. 2017SmoothGrad: removing noise by adding noisecited by 1,109
  5. 2017Interpretable Explanations of Black Boxes by Meaningful Perturbationcited by 612
  6. 2017Interpretation of Neural Networks is Fragilecited by 444
  7. 2019Fooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methodscited by 443
  8. 2019The many Shapley values for model explanationcited by 344
  9. 2018On the Robustness of Interpretability Methodscited by 314
  10. 2019On the (In)fidelity and Sensitivity for Explanationscited by 310
  11. 2019Explaining individual predictions when features are dependent: More accurate approximations to Shapley valuescited by 208
  12. 2019Feature relevance quantification in explainable AI: A causal problemcited by 167
15 more
  1. 2018Local Rule-Based Explanations of Black Box Decision Systemscited by 166
  2. 2019Explainable Machine Learning for Scientific Insights and Discoveriescited by 166
  3. 2020Improving KernelSHAP: Practical Shapley Value Estimation via Linear Regressioncited by 154
  4. 2017What do we need to build explainable AI systems for the medical domain?cited by 129
  5. 2020True to the Model or True to the Data?cited by 126
  6. 2019Asymmetric Shapley values: incorporating causal knowledge into model-agnostic explainabilitycited by 114
  7. 2018Model Agnostic Supervised Local Explanationscited by 83
  8. 2020Evaluations and Methods for Explanation through Robustness Analysiscited by 51
  9. 2019Measurable Counterfactual Local Explanations for Any Classifiercited by 48
  10. 2020OptiLIME: Optimized LIME Explanations for Diagnostic Computer Algorithmscited by 30
  11. 2020Improving LIME Robustness with Smarter Locality Samplingcited by 12
  12. 2020LimeOut: An Ensemble Approach To Improve Process Fairnesscited by 11
  13. 2019LoRMIkA: Local rule-based model interpretability with k-optimal associationscited by 10
  14. 2020MeLIME: Meaningful Local Explanation for Machine Learning Modelscited by 9
  15. 2019Local Interpretation Methods to Machine Learning Using the Domain of the Feature Spacecited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 5; 2022: 8; 2023: 5; 2024: 4; 2025: 1; 2026: 2.

25 papers.

Sort:

  1. 2022CRAFT: Concept Recursive Activation FacTorization for Explainabilitycited by 178
  2. 2022Post-Hoc Explanations Fail to Achieve their Purpose in Adversarial Contextscited by 53
  3. 2023Adversarial attacks and defenses in explainable artificial intelligence: A surveycited by 48
  4. 2022Don't Lie to Me! Robust and Efficient Explainability with Verified Perturbation Analysiscited by 38
  5. 2022Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measurecited by 33
  6. 2021RKHS-SHAP: Shapley Values for Kernel Methodscited by 30
  7. 2023Explaining the Uncertain: Stochastic Shapley Values for Gaussian Process Modelscited by 19
  8. 2022Unfooling Perturbation-Based Post Hoc Explainerscited by 12
  9. 2021When and How to Fool Explainable Models (and Humans) with Adversarial Examplescited by 10
  10. 2023Manifold Restricted Interventional Shapley Valuescited by 9
  11. 2022Explaining Preferences with Shapley Valuescited by 6
  12. 2021Exact Shapley Values for Local and Model-True Explanations of Decision Tree Ensemblescited by 5
13 more
  1. 2024Feature Accentuation: Revealing 'What' Features Respond to in Natural Imagescited by 5
  2. 2021Consistent Sufficient Explanations and Minimal Local Rules for explaining regression and classification modelscited by 4
  3. 2022Heterogeneous Treatment Effect with Trained Kernels of the Nadaraya-Watson Regressioncited by 4
  4. 2024Efficient and Accurate Explanation Estimation with Distribution Compressioncited by 4
  5. 2025Pixel-level Certified Explanations via Randomized Smoothingcited by 4
  6. 2021Attention-like feature explanation for tabular datacited by 3
  7. 2023Calculating and Visualizing Counterfactual Feature Importance Valuescited by 2
  8. 2022BENK: The Beran Estimator with Neural Kernels for Estimating the Heterogeneous Treatment Effectcited by 1
  9. 2023Explaining text classifiers through progressive neighborhood approximation with realistic samplescited by 1
  10. 2024Transparent Neighborhood Approximation for Text Classifier Explanationcited by 1
  11. 2024Dual feature-based and example-based explanation methodscited by 0
  12. 2026Empirical Characterization of Rationale Stability Under Controlled Perturbations for Explainable Pattern Recognitioncited by 0
  13. 2026Verified SHAP: Provable Bounds for Exact Shapley Values of Neural Networkscited by 0