Generating Tabular Data Using Heterogeneous Sequential Feature Forest Flow Matching

TL;DR

HS3F is a new method for generating synthetic tabular data that improves speed, quality, and robustness over previous flow-based models by sequentially generating features and using multinomial sampling for categorical variables.

Contribution

The paper introduces HS3F, a novel sequential feature generation approach that addresses speed and accuracy issues in existing flow-based tabular data generators.

Findings

HS3F produces higher quality synthetic data than FF.

HS3F is 21-27 times faster on datasets with many categorical variables.

HS3F shows increased robustness to flow initial condition variations.

Abstract

Privacy and regulatory constraints make data generation vital to advancing machine learning without relying on real-world datasets. A leading approach for tabular data generation is the Forest Flow (FF) method, which combines Flow Matching with XGBoost. Despite its good performance, FF is slow and makes errors when treating categorical variables as one-hot continuous features. It is also highly sensitive to small changes in the initial conditions of the ordinary differential equation (ODE). To overcome these limitations, we develop Heterogeneous Sequential Feature Forest Flow (HS3F). Our method generates data sequentially (feature-by-feature), reducing the dependency on noisy initial conditions through the additional information from previously generated features. Furthermore, it generates categorical variables using multinomial sampling (from an XGBoost classifier) instead of flow matching, improving generation speed. We also use a Runge-Kutta 4th order (Rg4) ODE solver for improved performance over the Euler solver used in FF. Our experiments with 25 datasets reveal that HS3F produces higher quality and more diverse synthetic data than FF, especially for categorical variables. It also generates data 21-27 times faster for datasets with $\geq20%$ categorical variables. HS3F further demonstrates enhanced robustness to affine transformation in flow ODE initial conditions compared to FF. This study not only validates the HS3F but also unveils promising new strategies to advance generative models.