TILBench: A Systematic Benchmark for Tabular Imbalanced Learning Across Data Regimes

2026-05-14

AI Summary

This study addresses the lack of systematic empirical analysis on the performance, robustness, and scalability of imbalanced learning methods across diverse tabular data characteristics. To this end, we introduce TILBench, a large-scale benchmark comprising 57 heterogeneous tabular datasets and over 40 representative algorithms—including resampling, cost-sensitive learning, and ensemble techniques. Through more than 200,000 controlled experiments conducted within a unified framework, we present the first comprehensive comparison of multiclass imbalanced learning approaches. Our findings reveal that no single method consistently outperforms others across all scenarios; instead, the effectiveness of each approach is highly dependent on dataset properties and computational constraints. Based on these empirical insights, we provide a practical, evidence-based guideline to inform algorithm selection in real-world applications.

Abstract

Imbalanced learning remains a fundamental challenge in tabular data applications. Despite decades of research and numerous proposed algorithms, a systematic empirical understanding of how different imbalanced learning methods behave across diverse data characteristics is still lacking. In particular, it remains unclear how different method families compare in predictive performance, robustness under varying data characteristics, and computational scalability. In this work, we present Tabular Imbalanced Learning Benchmark (TILBench), a large-scale empirical benchmark for tabular imbalanced learning. TILBench evaluates more than 40 representative algorithms across 57 diverse tabular datasets, resulting in over 200000 controlled experiments across a wide range of data characteristics. Our findings show that no single method consistently dominates across all settings; instead, the effectiveness of imbalanced learning methods depends strongly on dataset characteristics and computational constraints. Based on these findings, we provide practical recommendations for selecting appropriate methods in real-world applications.

Problem

• imbalanced learning
• tabular data
• systematic benchmark
• data regimes
• algorithm evaluation

Innovation

• imbalanced learning
• tabular data
• systematic benchmark
• algorithm evaluation
• empirical study