Interpretable Quantile Regression by Optimal Decision Trees

arXiv cs.LG / 4/24/2026

📰 NewsIdeas & Deep AnalysisModels & Research

共有:

Key Points

The paper introduces a novel learning method that produces an ensemble of optimal quantile regression trees aimed at improving both interpretability and robustness in machine learning models.
Unlike approaches that assume a specific form for the target distribution, the method generates predictions for the full conditional distribution of the outcome by using quantile regression.
The resulting trees are designed to remain interpretable, helping end users understand and build trust in the model’s outputs.
The authors claim algorithmic efficiency is not significantly compromised versus training a single quantile regression tree, despite learning multiple optimal trees.

Abstract

The field of machine learning is subject to an increasing interest in models that are not only accurate but also interpretable and robust, thus allowing their end users to understand and trust AI systems. This paper presents a novel method for learning a set of optimal quantile regression trees. The advantages of this method are that (1) it provides predictions about the complete conditional distribution of a target variable without prior assumptions on this distribution; (2) it provides predictions that are interpretable; (3) it learns a set of optimal quantile regression trees without compromising algorithmic efficiency compared to learning a single tree.