Interval-valued SHAP in Tree-Based Models
In this paper, we propose an imprecise Dirichlet model (IDM) based method to analyze the robustness of Shapley values in decision trees and random forests.
ProofPaper ↗
Key points
- Shapley values are among the most popular feature-attribution explanations.
- Efficient approaches for computing/estimating Shapley values for tree-based models, which are state-of-the-art for tabular data sets, have been developed.
- Technically, it is done by quantifying and analyzing the interval-valued Shapley values when a few unannotated instances are randomly introduced to the leaves of the trees.
- We then present various case studies and experiments to illustrate the behaviour of the proposed interval-valued Shapley values and their applications in debiasing uninformative features.
Sources (1)
- [1]Interval-valued SHAP in Tree-Based ModelsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 01:39 PM
In this paper, we propose an imprecise Dirichlet model (IDM) based method to analyze the robustness of Shapley values in decision trees and random forests.
Shapley values are among the most popular feature-attribution explanations.
Extractive summary: sentences quoted from the sources.