Does Training with Synthetic Data Truly Protect Privacy?
Keywords: ML privacy, membership inference
TL;DR: We rigorously evaluate privacy leakage across various methods based on training with synthetic data, and none of these methods achieve a better trade-off than the differential privacy baselines.
Abstract: As synthetic data becomes increasingly popular in machine learning tasks, numerous methods—without formal differential privacy guarantees—use synthetic data for training. These methods often claim, either explicitly or implicitly, to protect the privacy of the original training data.
In this work, we explore four different training paradigms—coreset selection, dataset distillation, data-free knowledge distillation, and synthetic data generated from diffusion models. While all these methods utilize synthetic data for training, they lead to vastly different conclusions regarding privacy preservation. This highlights that empirical approaches to preserving data privacy require careful and rigorous evaluation; otherwise, they risk providing a false sense of privacy.
Supplementary Material: zip
Primary Area: alignment, fairness, safety, privacy, and societal considerations
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 722
Loading