What type of study is this?

This is a Quantitative Study study.

October 1, 2025Open Access

Probabilistic vs Deep Generative Models: A Fairness Centred Evaluation of Synthetic Healthcare Tabular Data

View Full Paper

Ask AI

Bookmark

Authors

DADima R. AlattalBrunel University of London BDBarbara DraghiBrunel University of London PMPuja MylesBrighton and Sussex Medical School

Discussion

Member takes

Implication

This evaluation compares probabilistic and deep generative models for generating synthetic healthcare data, revealing significant differences in fairness and utility.

Key Points

BayesBoost achieved superior fidelity and fairness, especially with Random Forest classifiers.
Deep generative models often compromised fairness for underrepresented groups, indicating a need for adjustments.
VAE showed promise in fairness preservation but at a cost to fidelity and utility.
Assessment of synthetic healthcare data generation methods must prioritize fairness alongside fidelity.

Cite This Study

Alattal et al. (2025) studied this question.

synapsesocial.com/papers/68dd91cffe798ba2fc498cc1 https://doi.org/10.21203/rs.3.rs-7565139/v1

View Full Paper

Ask AI

Bookmark

Implication

This evaluation compares probabilistic and deep generative models for generating synthetic healthcare data, revealing significant differences in fairness and utility.

Probabilistic vs Deep Generative Models: A Fairness Centred Evaluation of Synthetic Healthcare Tabular Data

Discussion

Member takes

Implication

Key Points

Cite This Study

Probabilistic vs Deep Generative Models: A Fairness Centred Evaluation of Synthetic Healthcare Tabular Data

Discussion

Member takes

Implication

Key Points

Cite This Study

Also Consider

Also Consider