On the Tractability of SHAP Explanations

DOI10.1613/JAIR.1.13283MaRDI QIDQ5094036zbMATH OpenOpenAlexFDO

Authors Guy Van den Broeck, Anton Lykov, Maximilian Schleich, Dan Suciu

Publication date 2 August 2022

Published in Journal of Artificial Intelligence Research (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2009.08634

zbMATH Keywords

machine learning

Mathematics Subject Classification ID

Artificial intelligence (68Txx)

Recommendations

• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Efficient computation of shap explanation scores for neural network classifiers via knowledge compilation
• A comparative study of methods for estimating model-agnostic Shapley value explanations
• On the failings of Shapley values for explainability

Cites work

• scientific article; zbMATH DE number 44384 (Why is no real title available?)
• scientific article; zbMATH DE number 1946853 (Why is no real title available?)
• A Tractable and Expressive Class of Marginal Contribution Nets and Its Applications
• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Graph-Based Algorithms for Boolean Function Manipulation
• On probabilistic inference by weighted model counting
• The Complexity of Counting Cuts and of Computing the Probability that a Graph is Connected
• Theory and Applications of Satisfiability Testing
• Treewidth in Verification: Local vs. Global

Cited in

(12)

• Using analogical proportions for explanations
• On the failings of Shapley values for explainability
• Efficient Shapley performance attribution for least-squares regression
• Explaining robust classification through prime implicants
• A comparative study of methods for estimating model-agnostic Shapley value explanations
• A \(k\)-additive Choquet integral-based approach to approximate the SHAP values for local interpretability in machine learning
• On marginal feature attributions of tree-based models
• Knowledge compilation. Preface.
• Explanation of pseudo-Boolean functions using cooperative game theory and prime implicants
• Score-based explanations in data management and machine learning: an answer-set programming approach to counterfactual analysis
• Feature necessity \& relevancy in ML classifier explanations
• Efficient computation of shap explanation scores for neural network classifiers via knowledge compilation

Describes a project that uses

Uses Software

• shap

This page was built for publication: On the Tractability of SHAP Explanations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5094036)