# Shapley explainability on the data manifold

arXiv: https://arxiv.org/abs/2006.01272 · HTML: https://mlsanity.com/paper/2006.01272/

## Key facts

- Cited by 88 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 1 Jun 2020.

## Built on by (most cited first, 30 of 88)

- [FastSHAP: Real-Time Shapley Value Estimation](https://mlsanity.com/paper/2107.07436/) (2021)
- [The Shapley Value in Machine Learning](https://mlsanity.com/paper/2202.05594/) (2022)
- [Explainable and Interpretable Multimodal Large Language Models: A Comprehensive Survey](https://mlsanity.com/paper/2412.02104/) (2024)
- [Learning to Estimate Shapley Values with Vision Transformers](https://mlsanity.com/paper/2206.05282/) (2022)
- [shapiq: Shapley Interactions for Machine Learning](https://mlsanity.com/paper/2410.01649/) (2024)
- [WeightedSHAP: analyzing and improving Shapley based feature attributions](https://mlsanity.com/paper/2209.13429/) (2022)
- [Guidelines and Evaluation of Clinical Explainable AI in Medical Image Analysis](https://mlsanity.com/paper/2202.10553/) (2022)
- [The Inadequacy of Shapley Values for Explainability](https://mlsanity.com/paper/2302.08160/) (2023)
- [Fast TreeSHAP: Accelerating SHAP Value Computation for Trees](https://mlsanity.com/paper/2109.09847/) (2021)
- [Explaining Reinforcement Learning with Shapley Values](https://mlsanity.com/paper/2306.05810/) (2023)
- [Counterfactual Shapley Additive Explanations](https://mlsanity.com/paper/2110.14270/) (2021)
- [CoRTX: Contrastive Framework for Real-time Explanation](https://mlsanity.com/paper/2303.02794/) (2023)
- [Using Shapley Values and Variational Autoencoders to Explain Predictive Models with Dependent Mixed Features](https://mlsanity.com/paper/2111.13507/) (2021)
- [On the Robustness of Removal-Based Feature Attributions](https://mlsanity.com/paper/2306.07462/) (2023)
- [The Manifold Hypothesis for Gradient-Based Explanations](https://mlsanity.com/paper/2206.07387/) (2022)
- [A Comparative Study of Methods for Estimating Conditional Shapley Values and When to Use Them](https://mlsanity.com/paper/2305.09536/) (2023)
- [Locality Alignment Improves Vision-Language Models](https://mlsanity.com/paper/2410.11087/) (2024)
- [Exact Computation of Any-Order Shapley Interactions for Graph Neural Networks](https://mlsanity.com/paper/2501.16944/) (2025)
- [Interpreting Representation Quality of DNNs for 3D Point Cloud Processing](https://mlsanity.com/paper/2111.03549/) (2021)
- [Don't be fooled: label leakage in explanation methods and the importance of their quantitative evaluation](https://mlsanity.com/paper/2302.12893/) (2023)
- [Can We Faithfully Represent Masked States to Compute Shapley Values on a DNN?](https://mlsanity.com/paper/2105.10719/) (2021)
- [Interpretability Guarantees with Merlin-Arthur Classifiers](https://mlsanity.com/paper/2206.00759/) (2022)
- [Incremental Permutation Feature Importance (iPFI): Towards Online Explanations on Data Streams](https://mlsanity.com/paper/2209.01939/) (2022)
- [Manifold Restricted Interventional Shapley Values](https://mlsanity.com/paper/2301.04041/) (2023)
- [Risks, Causes, and Mitigations of Widespread Deployments of Large Language Models (LLMs): A Survey](https://mlsanity.com/paper/2408.04643/) (2024)
- [Local Explanation of Dialogue Response Generation](https://mlsanity.com/paper/2106.06528/) (2021)
- [Fool SHAP with Stealthily Biased Sampling](https://mlsanity.com/paper/2205.15419/) (2022)
- [SHAP-XRT: The Shapley Value Meets Conditional Independence Testing](https://mlsanity.com/paper/2207.07038/) (2022)
- [A Survey of Privacy-Preserving Model Explanations: Privacy Risks, Attacks, and Countermeasures](https://mlsanity.com/paper/2404.00673/) (2024)
- [Exact Shapley Values for Local and Model-True Explanations of Decision Tree Ensembles](https://mlsanity.com/paper/2112.10592/) (2021)

## Abstract (arXiv)

Explainability in AI is crucial for model development, compliance with regulation, and providing operational nuance to predictions. The Shapley framework for explainability attributes a model's predictions to its input features in a mathematically principled and model-agnostic way. However, general implementations of Shapley explainability make an untenable assumption: that the model's features are uncorrelated. In this work, we demonstrate unambiguous drawbacks of this assumption and develop two solutions to Shapley explainability that respect the data manifold. One solution, based on generative modelling, provides flexible access to data imputations; the other directly learns the Shapley value-function, providing performance and stability at the cost of flexibility. While "off-manifold" Shapley values can (i) give rise to incorrect explanations, (ii) hide implicit model dependence on sensitive attributes, and (iii) lead to unintelligible explanations in higher-dimensional data, on-manifold explainability overcomes these problems.