A comparative study of methods for estimating model-agnostic Shapley value explanations

DOI10.1007/S10618-024-01016-ZMaRDI QIDQ6609084zbMATH OpenFDO

Authors Lars Henry Berge Olsen, Ingrid Kristine Glad, Martin Jullum, Kjersti Aas

Publication date 20 September 2024

Published in Data Mining and Knowledge Discovery (Search for Journal in Brave)

Copyright license Creative Commons Attribution 4.0 International

Full work available at URL https://doi.org/10.1007/S10618-024-01016-Z

zbMATH Keywords

explainable artificial intelligencefeature dependencefeature importancemodel-agnostic explanationprediction explanationShapley values

Mathematics Subject Classification ID

Nonparametric estimation (62G05) General topics in artificial intelligence (68T01) Learning and adaptive systems in artificial intelligence (68T05) Applications of game theory (91A80)

Recommendations

• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Explaining predictive models using Shapley values and non-parametric vine copulas
• On the Tractability of SHAP Explanations
• A k-additive Choquet integral-based approach to approximate the SHAP values for local interpretability in machine learning
• On Shapley Value for Measuring Importance of Dependent Inputs

Cites work

• A decision-theoretic generalization of on-line learning and an application to boosting
• A mixture of generalized hyperbolic distributions
• A Simplex Method for Function Minimization
• An efficient explanation of individual classifications using game theory
• Analysis of regression in game theory approach
• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Explaining predictive models using Shapley values and non-parametric vine copulas
• Generalized additive models. An introduction with R.
• scientific article; zbMATH DE number 4027220 (Why is no real title available?)
• scientific article; zbMATH DE number 7626724 (Why is no real title available?)
• scientific article; zbMATH DE number 7625196 (Why is no real title available?)
• scientific article; zbMATH DE number 3078997 (Why is no real title available?)
• Least angle regression. (With discussion)
• Low‐Rank Scale‐Invariant Tensor Product Smooths for Generalized Additive Mixed Models
• Mixtures of generalized hyperbolic distributions and mixtures of skew-t distributions for model-based clustering with incomplete data
• Note on the multivariate Burr's distribution
• On the Complexity of Cooperative Solution Concepts
• Quantitative risk management. Concepts, techniques and tools
• Random forests
• Sobol' indices and Shapley value
• The Elements of Statistical Learning
• THE POPULATION FREQUENCIES OF SPECIES AND THE ESTIMATION OF POPULATION PARAMETERS
• The Shapley value for cooperative games under precedence constraints

Cited in

(8)

• On the Tractability of SHAP Explanations
• Variance reduced Shapley value estimation for trustworthy data valuation
• A k-additive Choquet integral-based approach to approximate the SHAP values for local interpretability in machine learning
• Efficient computation of shap explanation scores for neural network classifiers via knowledge compilation
• On the failings of Shapley values for explainability
• Efficient Shapley performance attribution for least-squares regression
• On marginal feature attributions of tree-based models
• Interpretable machine learning for survival analysis

This page was built for publication: A comparative study of methods for estimating model-agnostic Shapley value explanations

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6609084)