A Comparative Analysis of Interpretable Machine Learning Methods

2026-01-01

arXiv.org

AI Summary

This study addresses the lack of systematic evaluation of intrinsically interpretable machine learning methods on real-world tabular data, particularly overlooking how data structural characteristics influence model performance. For the first time, it comprehensively benchmarks 16 interpretable models—including linear models, decision trees, Explainable Boosting Machines (EBM), symbolic regression, Generalized Optimal Sparse Decision Trees (GOSDT), and Interpretable Graph Additive Neural Networks (IGANN)—across 216 real tabular datasets stratified by dimensionality, sample size, linearity, and class imbalance. The evaluation assesses predictive accuracy, training efficiency, and robustness under distributional shift. Results show that EBM achieves the best performance in regression tasks, while symbolic regression and IGANN excel in highly nonlinear settings. GOSDT, however, proves sensitive to class imbalance. This work provides practitioners with an empirically grounded, data-characteristic-driven guide for selecting interpretable models.

Technology Category

Machine Learning: Transparent, Interpretable, Explainable MLNatural Language Processing: Interpretability, Analysis, and Evaluation of NLP ModelsCognitive Modeling & Cognitive Systems: Symbolic Representations

Application Category

Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semanticsUser Modeling, Personalization and Recommendation: Explainable and interpretable methods for personalizationSearch and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

In recent years, Machine Learning (ML) has seen widespread adoption across a broad range of sectors, including high-stakes domains such as healthcare, finance, and law. This growing reliance has raised increasing concerns regarding model interpretability and accountability, particularly as legal and regulatory frameworks place tighter constraints on using black-box models in critical applications. Although interpretable ML has attracted substantial attention, systematic evaluations of inherently interpretable models, especially for tabular data, remain relatively scarce and often focus primarily on aggregated performance outcomes. To address this gap, we present a large-scale comparative evaluation of 16 inherently interpretable methods, ranging from classical linear models and decision trees to more recent approaches such as Explainable Boosting Machines (EBMs), Symbolic Regression (SR), and Generalized Optimal Sparse Decision Trees (GOSDT). Our study spans 216 real-world tabular datasets and goes beyond aggregate rankings by stratifying performance according to structural dataset characteristics, including dimensionality, sample size, linearity, and class imbalance. In addition, we assess training time and robustness under controlled distributional shifts. Our results reveal clear performance hierarchies, especially for regression tasks, where EBMs consistently achieve strong predictive accuracy. At the same time, we show that performance is highly context-dependent: SR and Interpretable Generalized Additive Neural Networks (IGANNs) perform particularly well in non-linear regimes, while GOSDT models exhibit pronounced sensitivity to class imbalance. Overall, these findings provide practical guidance for practitioners seeking a balance between interpretability and predictive performance, and contribute to a deeper empirical understanding of interpretable modeling for tabular data.

Problem

• Interpretable Machine Learning
• Tabular Data
• Model Interpretability
• Comparative Evaluation
• Black-box Models

Innovation

• Interpretable Machine Learning
• Tabular Data
• Explainable Boosting Machines
• Symbolic Regression
• Generalized Optimal Sparse Decision Trees