Beyond the convexity assumption: Realistic tabular data generation under quantifier-free real linear constraints

TL;DR

This paper introduces the Disjunctive Refinement Layer (DRL), a novel method that ensures deep generative models produce realistic, constraint-compliant tabular data, significantly improving data quality and downstream task performance.

Contribution

The paper presents DRL, the first layer enabling deep models to automatically satisfy complex linear constraints, including non-convex and disconnected spaces, in tabular data generation.

Findings

DRL guarantees constraint satisfaction in generated data.

DRL eliminates violations of user-defined constraints.

Improves downstream task metrics by up to 21.4% F1-score and 20.9% AUC.

Abstract

Synthetic tabular data generation has traditionally been a challenging problem due to the high complexity of the underlying distributions that characterise this type of data. Despite recent advances in deep generative models (DGMs), existing methods often fail to produce realistic datapoints that are well-aligned with available background knowledge. In this paper, we address this limitation by introducing Disjunctive Refinement Layer (DRL), a novel layer designed to enforce the alignment of generated data with the background knowledge specified in user-defined constraints. DRL is the first method able to automatically make deep learning models inherently compliant with constraints as expressive as quantifier-free linear formulas, which can define non-convex and even disconnected spaces. Our experimental analysis shows that DRL not only guarantees constraint satisfaction but also improves efficacy in downstream tasks. Notably, when applied to DGMs that frequently violate constraints, DRL eliminates violations entirely. Further, it improves performance metrics by up to 21.4% in F1-score and 20.9% in Area Under the ROC Curve, thus demonstrating its practical impact on data generation.