Classical methods for modern applications: synthetic tabular data generation using lightweight discrete response continuous latent variable model
Abstract
In this paper, we present a lightweight algorithm for modelling tabular data containing both continuous and discrete variables. The method uses empirical marginal distributions and a selected quantile function to transform the data into a continuous latent space, where a distribution of choice, e.g. a multivariate Gaussian, can be efficiently estimated and applied. While this approach is closely related to classical well-established machine learning methods, we present a novel adaptation that fits well within the contemporary computational ecosystem. Unlike in the case of deep architectures, it results in a simple implementation with few dependencies, robust training/estimation, as well as efficient sampling. The transformation preserves complex marginal distributions and correlations, providing high-fidelity models, especially for data with unimodal distributions. One of the key characteristics of the proposed approach is its continuous representation of discrete variables. While highly desirable, this introduces a significant challenge, since a single observation corresponds to a region rather than a point in the latent space. We show that this ambiguity naturally leads to a Monte Carlo expectation-maximisation (MCEM) formulation, whose auxiliary objective can be expressed as a weighted latent-space likelihood. This, in turn, allows conventional maximum-likelihood estimators to be used within an efficient and robust optimisation procedure.
Related
- A Modeling Method for Continuous Active Synthetic Sound During Dynamic Operating Condition Transitions in Electric Vehicles
- DagFlow AI: Agentic Synthetic Data Generation for Personalized Statistics Learning
- Synthetic Data Augmentation for Functional Data: Wasserstein Risk Bounds and Diffusion-Based Conditional Generation
- Verifier-Guided Synthetic Augmentation for 3D Human Shape Generation
- An Energy-Driven Framework for Privacy-Aware Synthetic Data Generation
BibTeX
@misc{barzdajn2026classical,
title = {{Classical methods for modern applications: synthetic tabular data generation using lightweight discrete response continuous latent variable model}},
author = {Bartosz Barzdajn and Mateusz Krzysztoń},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.21674},
eprint = {2610.21674},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.21674},
}
RIS
TY - UNPB
TI - Classical methods for modern applications: synthetic tabular data generation using lightweight discrete response continuous latent variable model
AU - Bartosz Barzdajn
AU - Mateusz Krzysztoń
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.21674
UR - https://arxiv.org/abs/2610.21674
AB - In this paper, we present a lightweight algorithm for modelling tabular data containing both continuous and discrete variables. The method uses empirical marginal distributions and a selected quantile function to transform the data into a continuous latent space, where a distribution of choice, e.g. a multivariate Gaussian, can be efficiently estimated and applied. While this approach is closely related to classical well-established machine learning methods, we present a novel adaptation that fits well within the contemporary computational ecosystem. Unlike in the case of deep architectures, it results in a simple implementation with few dependencies, robust training/estimation, as well as efficient sampling. The transformation preserves complex marginal distributions and correlations, providing high-fidelity models, especially for data with unimodal distributions. One of the key characteristics of the proposed approach is its continuous representation of discrete variables. While highly desirable, this introduces a significant challenge, since a single observation corresponds to a region rather than a point in the latent space. We show that this ambiguity naturally leads to a Monte Carlo expectation-maximisation (MCEM) formulation, whose auxiliary objective can be expressed as a weighted latent-space likelihood. This, in turn, allows conventional maximum-likelihood estimators to be used within an efficient and robust optimisation procedure.
ER -