On marginal feature attributions of tree-based models

DOI10.3934/FODS.2024021MaRDI QIDQ6620135zbMATH OpenFDO

Authors Khashayar Filom, Alexey Miroshnikov, Konstandinos Kotsiopoulos, Arjun Ravi Kannan

Publication date 16 October 2024

Published in Foundations of Data Science (Search for Journal in Brave)

zbMATH Keywords

tree ensemblescooperative gamegame valueML interpretability

Mathematics Subject Classification ID

Trees (05C05) Cooperative games (91A12) Applications of game theory (91A80) Combinatorial identities, bijective combinatorics (05A19) (n)-person games, (n>2) (91A06) General topics in artificial intelligence (68T01) Algorithmic game theory and complexity (91A68)

Recommendations

• On the Tractability of SHAP Explanations
• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Interpreting deep learning models with marginal attribution by conditioning on quantiles
• A comparative study of methods for estimating model-agnostic Shapley value explanations

Cites work

• scientific article; zbMATH DE number 3614541 (Why is no real title available?)
• scientific article; zbMATH DE number 3078997 (Why is no real title available?)
• A two-step Shapley value for cooperative games with coalition structures
• Cooperative games with coalition structures
• Explaining individual predictions when features are dependent: more accurate approximations to Shapley values
• Greedy function approximation: A gradient boosting machine.
• Measuring dependence powerfully and equitably
• Monotonic solutions of cooperative games
• Random forests
• Supervised Machine Learning Techniques: An Overview with Applications to Banking
• The Elements of Statistical Learning
• Theory of probability and random processes.

This page was built for publication: On marginal feature attributions of tree-based models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6620135)