Synthetic Tabular Data: Methods, Attacks and Defenses

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cormode, Graham, Maddock, Samuel, Ullah, Enayat, Gade, Shripad
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908396485083136
author Cormode, Graham
Maddock, Samuel
Ullah, Enayat
Gade, Shripad
author_facet Cormode, Graham
Maddock, Samuel
Ullah, Enayat
Gade, Shripad
contents Synthetic data is often positioned as a solution to replace sensitive fixed-size datasets with a source of unlimited matching data, freed from privacy concerns. There has been much progress in synthetic data generation over the last decade, leveraging corresponding advances in machine learning and data analytics. In this survey, we cover the key developments and the main concepts in tabular synthetic data generation, including paradigms based on probabilistic graphical models and on deep learning. We provide background and motivation, before giving a technical deep-dive into the methodologies. We also address the limitations of synthetic data, by studying attacks that seek to retrieve information about the original sensitive data. Finally, we present extensions and open problems in this area.
format Preprint
id arxiv_https___arxiv_org_abs_2506_06108
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Synthetic Tabular Data: Methods, Attacks and Defenses
Cormode, Graham
Maddock, Samuel
Ullah, Enayat
Gade, Shripad
Machine Learning
Cryptography and Security
Synthetic data is often positioned as a solution to replace sensitive fixed-size datasets with a source of unlimited matching data, freed from privacy concerns. There has been much progress in synthetic data generation over the last decade, leveraging corresponding advances in machine learning and data analytics. In this survey, we cover the key developments and the main concepts in tabular synthetic data generation, including paradigms based on probabilistic graphical models and on deep learning. We provide background and motivation, before giving a technical deep-dive into the methodologies. We also address the limitations of synthetic data, by studying attacks that seek to retrieve information about the original sensitive data. Finally, we present extensions and open problems in this area.
title Synthetic Tabular Data: Methods, Attacks and Defenses
topic Machine Learning
Cryptography and Security
url https://arxiv.org/abs/2506.06108