Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Amalgam: Hybrid LLM-PGM Synthesis Algorithm for Accuracy and Realism

Antheas Kapenekakis, Bent Thomsen, Katja Hose, Michele Albano

Abstract

To generate synthetic datasets, e.g., in domains such as healthcare, the literature proposes approaches of two main types: Probabilistic Graphical Models (PGMs) and Deep Learning models, such as LLMs. While PGMs produce synthetic data that can be used for advanced analytics, they do not support complex schemas and datasets. LLMs on the other hand, support complex schemas but produce skewed dataset distributions, which are less useful for advanced analytics. In this paper, we therefore present Amalgam, a hybrid LLM-PGM data synthesis algorithm supporting both advanced analytics, realism, and tangible privacy properties. We show that Amalgam synthesizes data with an average 91 % $χ^2 P$ value and scores 3.8/5 for realism using our proposed metric, where state-of-the-art is 3.3 and real data is 4.7.

Amalgam: Hybrid LLM-PGM Synthesis Algorithm for Accuracy and Realism

Abstract

To generate synthetic datasets, e.g., in domains such as healthcare, the literature proposes approaches of two main types: Probabilistic Graphical Models (PGMs) and Deep Learning models, such as LLMs. While PGMs produce synthetic data that can be used for advanced analytics, they do not support complex schemas and datasets. LLMs on the other hand, support complex schemas but produce skewed dataset distributions, which are less useful for advanced analytics. In this paper, we therefore present Amalgam, a hybrid LLM-PGM data synthesis algorithm supporting both advanced analytics, realism, and tangible privacy properties. We show that Amalgam synthesizes data with an average 91 % value and scores 3.8/5 for realism using our proposed metric, where state-of-the-art is 3.3 and real data is 4.7.

Paper Structure

This paper contains 28 sections, 1 equation, 3 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Amalgam Algorithm
  4. Structure Learning
  5. Sampling
  6. Realism Evaluation
  7. Experiments
  8. Comparison with State-of-the-Art
  9. LLM Comparison and Efficiency
  10. Hardware Setup
  11. Implementation
  12. Results
  13. Discussion
  14. Conclusion
  15. Acknowledgments
  16. ...and 13 more sections

Figures (3)

  • Figure 1: Comparison of Amalgam synthesis results across datasets versus MARE, Real Data.
  • Figure 2: Comparison of Amalgam synthesis for different LLMs on MIMIC-IV Admissions.
  • Figure 3: Energy usage comparison across different LLMs on MIMIC-IV Admissions dataset.