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

2025-02-25

AI Summary

Synthetic tabular data generation often fails to satisfy complex linear background knowledge—specifically, quantifier-free real linear constraints that are non-convex or disconnected. Method: This paper introduces the Discrete Refinement Layer (DRL), the first differentiable module enabling deep generative models to enforce such constraints automatically and intrinsically, without dependence on quantization schemes. DRL integrates differentiable constraint encoding with real linear logic reasoning and gradient-based optimization, ensuring seamless compatibility with mainstream generative architectures. Results: DRL guarantees zero constraint violations while preserving data fidelity. Empirical evaluation demonstrates substantial improvements in downstream task performance: up to +21.4% in F1-score and +20.9% in AUC, significantly enhancing consistency between generated data and domain-specific linear knowledge.

Technology Category

• Constraint Satisfaction and Optimization: Satisfiability Modulo Theories
• Natural Language Processing: Generation
• Machine Learning: Deep Generative Models & Autoencoders

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

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.

Problem

• Generate realistic synthetic tabular data
• Enforce alignment with background knowledge
• Comply with quantifier-free linear constraints

Innovation

• Disjunctive Refinement Layer
• quantifier-free linear formulas
• non-convex spaces