TabReD: Analyzing Pitfalls and Filling the Gaps in Tabular Deep Learning Benchmarks

TL;DR

This paper introduces TabReD, a collection of industry-grade tabular datasets, and analyzes how recent deep learning methods perform under real-world conditions like data drift and feature engineering, revealing different rankings than traditional benchmarks.

Contribution

The paper presents TabReD, a new benchmark dataset collection, and evaluates existing models under realistic industrial scenarios, highlighting the importance of time-based splits and feature considerations.

Findings

Time-based splits change model rankings.

Simple MLP and GBDT perform best on TabReD.

Current benchmarks may not reflect real-world challenges.

Abstract

Advances in machine learning research drive progress in real-world applications. To ensure this progress, it is important to understand the potential pitfalls on the way from a novel method's success on academic benchmarks to its practical deployment. In this work, we analyze existing tabular benchmarks and find two common characteristics of tabular data in typical industrial applications that are underrepresented in the datasets usually used for evaluation in the literature. First, in real-world deployment scenarios, distribution of data often changes over time. To account for this distribution drift, time-based train/test splits should be used in evaluation. However, popular tabular datasets often lack timestamp metadata to enable such evaluation. Second, a considerable portion of datasets in production settings stem from extensive data acquisition and feature engineering pipelines. This can have an impact on the absolute and relative number of predictive, uninformative, and correlated features compared to academic datasets. In this work, we aim to understand how recent research advances in tabular deep learning transfer to these underrepresented conditions. To this end, we introduce TabReD -- a collection of eight industry-grade tabular datasets. We reassess a large number of tabular ML models and techniques on TabReD. We demonstrate that evaluation on time-based data splits leads to different methods ranking, compared to evaluation on random splits, which are common in current benchmarks. Furthermore, simple MLP-like architectures and GBDT show the best results on the TabReD datasets, while other methods are less effective in the new setting.