UbiQTree: Uncertainty Quantification in XAI with Tree Ensembles

TL;DR

UbiQTree introduces a method to decompose and quantify uncertainty in SHAP explanations for tree ensemble models, enhancing interpretability and reliability in high-stakes domains.

Contribution

It presents a novel approach combining Dempster-Shafer theory and Dirichlet processes to separate aleatoric and epistemic uncertainties in SHAP values.

Findings

Features with high SHAP values may have high epistemic uncertainty.

Tree ensembles, especially bagging, effectively quantify epistemic uncertainty.

Reducing epistemic uncertainty improves model reliability in critical applications.

Abstract

Explainable Artificial Intelligence (XAI) techniques, such as SHapley Additive exPlanations (SHAP), have become essential tools for interpreting complex ensemble tree-based models, especially in high-stakes domains such as healthcare analytics. However, SHAP values are usually treated as point estimates, which disregards the inherent and ubiquitous uncertainty in predictive models and data. This uncertainty has two primary sources: aleatoric and epistemic. The aleatoric uncertainty, which reflects the irreducible noise in the data. The epistemic uncertainty, which arises from a lack of data. In this work, we propose an approach for decomposing uncertainty in SHAP values into aleatoric, epistemic, and entanglement components. This approach integrates Dempster-Shafer evidence theory and hypothesis sampling via Dirichlet processes over tree ensembles. We validate the method across three real-world use cases with descriptive statistical analyses that provide insight into the nature of epistemic uncertainty embedded in SHAP explanations. The experimentations enable to provide more comprehensive understanding of the reliability and interpretability of SHAP-based attributions. This understanding can guide the development of robust decision-making processes and the refinement of models in high-stakes applications. Through our experiments with multiple datasets, we concluded that features with the highest SHAP values are not necessarily the most stable. This epistemic uncertainty can be reduced through better, more representative data and following appropriate or case-desired model development techniques. Tree-based models, especially bagging, facilitate the effective quantification of epistemic uncertainty.