Tree-based local explanations of machine learning model predictions, AraucanaXAI

TL;DR

This paper introduces a novel tree-based local explanation method for machine learning models that enhances interpretability while maintaining high fidelity, applicable to both classification and regression tasks, especially useful in high-stakes domains.

Contribution

The paper presents a new XAI approach that improves fidelity and handles non-linear boundaries, supporting both classification and regression, addressing interpretability challenges in complex models.

Findings

Improved explanation fidelity to original models

Supports both classification and regression tasks

Handles non-linear decision boundaries

Abstract

Increasingly complex learning methods such as boosting, bagging and deep learning have made ML models more accurate, but harder to understand and interpret. A tradeoff between performance and intelligibility is often to be faced, especially in high-stakes applications like medicine. In the present article we propose a novel methodological approach for generating explanations of the predictions of a generic ML model, given a specific instance for which the prediction has been made, that can tackle both classification and regression tasks. Advantages of the proposed XAI approach include improved fidelity to the original model, the ability to deal with non-linear decision boundaries, and native support to both classification and regression problems