# Towards Privacy-Preserving Relational Data Synthesis via Probabilistic Relational Models

arXiv: https://arxiv.org/abs/2409.04194 · HTML: https://mlsanity.com/paper/2409.04194/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 36 references that are in the map.
- First posted on arXiv 6 Sep 2024.

## Builds on (most cited first, 7 of 7)

- [Modeling Tabular data using Conditional GAN](https://mlsanity.com/paper/1907.00503/) (2019)
- [Synthetic Data for Deep Learning](https://mlsanity.com/paper/1909.11512/) (2019)
- [Lifted Variable Elimination: Decoupling the Operators from the Constraint Language](https://mlsanity.com/paper/1402.0565/) (2014)
- [PrivLava: Synthesizing Relational Data with Foreign Keys under Differential Privacy](https://mlsanity.com/paper/2304.04545/) (2023)
- [Colour Passing Revisited: Lifted Model Construction with Commutative Factors](https://mlsanity.com/paper/2309.11236/) (2023)
- [Efficient Detection of Exchangeable Factors in Factor Graphs](https://mlsanity.com/paper/2403.10167/) (2024)
- [Lifted Causal Inference in Relational Domains](https://mlsanity.com/paper/2403.10184/) (2024)

## Abstract (arXiv)

Probabilistic relational models provide a well-established formalism to combine first-order logic and probabilistic models, thereby allowing to represent relationships between objects in a relational domain. At the same time, the field of artificial intelligence requires increasingly large amounts of relational training data for various machine learning tasks. Collecting real-world data, however, is often challenging due to privacy concerns, data protection regulations, high costs, and so on. To mitigate these challenges, the generation of synthetic data is a promising approach. In this paper, we solve the problem of generating synthetic relational data via probabilistic relational models. In particular, we propose a fully-fledged pipeline to go from relational database to probabilistic relational model, which can then be used to sample new synthetic relational data points from its underlying probability distribution. As part of our proposed pipeline, we introduce a learning algorithm to construct a probabilistic relational model from a given relational database.