Fast Calculation of Feature Contributions in Boosting Trees

Zhongli Jiang, Min Zhang, Dabao Zhang
Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, PMLR 286:1859-1875, 2025.

Abstract

Recently, several fast algorithms have been proposed to decompose predicted value into Shapley values, enabling individualized feature contribution analysis in tree models. While such local decomposition offers valuable insights, it underscores the need for a global evaluation of feature contributions. Although coefficients of determination ($R^2$) allow for comparative assessment of individual features, individualizing $R^2$ is challenged by the underlying quadratic losses. To address this, we propose Q-SHAP, an efficient algorithm that reduces the computational complexity of calculating Shapley values for quadratic losses to polynomial time. Our extensive simulations show that Q-SHAP not only improves computational efficiency but also enhances the accuracy of feature-specific $R^2$ estimates.

Cite this Paper

BibTeX
@InProceedings{pmlr-v286-jiang25a, title = {Fast Calculation of Feature Contributions in Boosting Trees}, author = {Jiang, Zhongli and Zhang, Min and Zhang, Dabao}, booktitle = {Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence}, pages = {1859--1875}, year = {2025}, editor = {Chiappa, Silvia and Magliacane, Sara}, volume = {286}, series = {Proceedings of Machine Learning Research}, month = {21--25 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v286/main/assets/jiang25a/jiang25a.pdf}, url = {https://proceedings.mlr.press/v286/jiang25a.html}, abstract = {Recently, several fast algorithms have been proposed to decompose predicted value into Shapley values, enabling individualized feature contribution analysis in tree models. While such local decomposition offers valuable insights, it underscores the need for a global evaluation of feature contributions. Although coefficients of determination ($R^2$) allow for comparative assessment of individual features, individualizing $R^2$ is challenged by the underlying quadratic losses. To address this, we propose Q-SHAP, an efficient algorithm that reduces the computational complexity of calculating Shapley values for quadratic losses to polynomial time. Our extensive simulations show that Q-SHAP not only improves computational efficiency but also enhances the accuracy of feature-specific $R^2$ estimates.} }
Endnote
%0 Conference Paper %T Fast Calculation of Feature Contributions in Boosting Trees %A Zhongli Jiang %A Min Zhang %A Dabao Zhang %B Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence %C Proceedings of Machine Learning Research %D 2025 %E Silvia Chiappa %E Sara Magliacane %F pmlr-v286-jiang25a %I PMLR %P 1859--1875 %U https://proceedings.mlr.press/v286/jiang25a.html %V 286 %X Recently, several fast algorithms have been proposed to decompose predicted value into Shapley values, enabling individualized feature contribution analysis in tree models. While such local decomposition offers valuable insights, it underscores the need for a global evaluation of feature contributions. Although coefficients of determination ($R^2$) allow for comparative assessment of individual features, individualizing $R^2$ is challenged by the underlying quadratic losses. To address this, we propose Q-SHAP, an efficient algorithm that reduces the computational complexity of calculating Shapley values for quadratic losses to polynomial time. Our extensive simulations show that Q-SHAP not only improves computational efficiency but also enhances the accuracy of feature-specific $R^2$ estimates.
APA
Jiang, Z., Zhang, M. & Zhang, D.. (2025). Fast Calculation of Feature Contributions in Boosting Trees. Proceedings of the Forty-first Conference on Uncertainty in Artificial Intelligence, in Proceedings of Machine Learning Research 286:1859-1875 Available from https://proceedings.mlr.press/v286/jiang25a.html.

Related Material