A Closer Look at Deep Learning Methods on Tabular Datasets

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ye, Han-Jia, Liu, Si-Yang, Cai, Hao-Run, Zhou, Qi-Le, Zhan, De-Chuan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915602994561024
author Ye, Han-Jia
Liu, Si-Yang
Cai, Hao-Run
Zhou, Qi-Le
Zhan, De-Chuan
author_facet Ye, Han-Jia
Liu, Si-Yang
Cai, Hao-Run
Zhou, Qi-Le
Zhan, De-Chuan
contents Tabular data is prevalent across diverse domains in machine learning. With the rapid progress of deep tabular prediction methods, especially pretrained (foundation) models, there is a growing need to evaluate these methods systematically and to understand their behavior. We present an extensive study on TALENT, a collection of 300+ datasets spanning broad ranges of size, feature composition (numerical/categorical mixes), domains, and output types (binary, multi--class, regression). Our evaluation shows that ensembling benefits both tree-based and neural approaches. Traditional gradient-boosted trees remain very strong baselines, yet recent pretrained tabular models now match or surpass them on many tasks, narrowing--but not eliminating--the historical advantage of tree ensembles. Despite architectural diversity, top performance concentrates within a small subset of models, providing practical guidance for method selection. To explain these outcomes, we quantify dataset heterogeneity by learning from meta-features and early training dynamics to predict later validation behavior. This dynamics-aware analysis indicates that heterogeneity--such as the interplay of categorical and numerical attributes--largely determines which family of methods is favored. Finally, we introduce a two-level design beyond the 300 common-size datasets: a compact TALENT-tiny core (45 datasets) for rapid, reproducible evaluation, and a TALENT-extension suite targeting high-dimensional, many-class, and very large-scale settings for stress testing. In summary, these results offer actionable insights into the strengths, limitations, and future directions for improving deep tabular learning.
format Preprint
id arxiv_https___arxiv_org_abs_2407_00956
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle A Closer Look at Deep Learning Methods on Tabular Datasets
Ye, Han-Jia
Liu, Si-Yang
Cai, Hao-Run
Zhou, Qi-Le
Zhan, De-Chuan
Machine Learning
Tabular data is prevalent across diverse domains in machine learning. With the rapid progress of deep tabular prediction methods, especially pretrained (foundation) models, there is a growing need to evaluate these methods systematically and to understand their behavior. We present an extensive study on TALENT, a collection of 300+ datasets spanning broad ranges of size, feature composition (numerical/categorical mixes), domains, and output types (binary, multi--class, regression). Our evaluation shows that ensembling benefits both tree-based and neural approaches. Traditional gradient-boosted trees remain very strong baselines, yet recent pretrained tabular models now match or surpass them on many tasks, narrowing--but not eliminating--the historical advantage of tree ensembles. Despite architectural diversity, top performance concentrates within a small subset of models, providing practical guidance for method selection. To explain these outcomes, we quantify dataset heterogeneity by learning from meta-features and early training dynamics to predict later validation behavior. This dynamics-aware analysis indicates that heterogeneity--such as the interplay of categorical and numerical attributes--largely determines which family of methods is favored. Finally, we introduce a two-level design beyond the 300 common-size datasets: a compact TALENT-tiny core (45 datasets) for rapid, reproducible evaluation, and a TALENT-extension suite targeting high-dimensional, many-class, and very large-scale settings for stress testing. In summary, these results offer actionable insights into the strengths, limitations, and future directions for improving deep tabular learning.
title A Closer Look at Deep Learning Methods on Tabular Datasets
topic Machine Learning
url https://arxiv.org/abs/2407.00956