Enabling PSO-Secure Synthetic Data Sharing Using Diversity-Aware Diffusion Models

• URL: http://arxiv.org/abs/2506.17975v1
• Date: Sun, 22 Jun 2025 10:26:35 GMT
• Title: Enabling PSO-Secure Synthetic Data Sharing Using Diversity-Aware Diffusion Models
• Authors: Mischa Dombrowski, Bernhard Kainz,
• Abstract summary: We propose a generalisable framework for training diffusion models on personal data.<n>This leads to unpersonal synthetic datasets achieving performance within one percentage point of real-data models.
• Score: 7.202078342390581
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Synthetic data has recently reached a level of visual fidelity that makes it nearly indistinguishable from real data, offering great promise for privacy-preserving data sharing in medical imaging. However, fully synthetic datasets still suffer from significant limitations: First and foremost, the legal aspect of sharing synthetic data is often neglected and data regulations, such as the GDPR, are largley ignored. Secondly, synthetic models fall short of matching the performance of real data, even for in-domain downstream applications. Recent methods for image generation have focused on maximising image diversity instead of fidelity solely to improve the mode coverage and therefore the downstream performance of synthetic data. In this work, we shift perspective and highlight how maximizing diversity can also be interpreted as protecting natural persons from being singled out, which leads to predicate singling-out (PSO) secure synthetic datasets. Specifically, we propose a generalisable framework for training diffusion models on personal data which leads to unpersonal synthetic datasets achieving performance within one percentage point of real-data models while significantly outperforming state-of-the-art methods that do not ensure privacy. Our code is available at https://github.com/MischaD/Trichotomy.

Related papers

• Improving Noise Efficiency in Privacy-preserving Dataset Distillation [59.57846442477106]
  We introduce a novel framework that decouples sampling from optimization for better convergence and improves signal quality.<n>On CIFAR-10, our method achieves a textbf10.0% improvement with 50 images per class and textbf8.3% increase with just textbfone-fifth the distilled set size of previous state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-08-03T13:15:52Z)
• SMOTE-DP: Improving Privacy-Utility Tradeoff with Synthetic Data [13.699107354397286]
  We show that, with the right mechanism of synthetic data generation, we can achieve strong privacy protection without significant utility loss.<n>We prove in theory and through empirical demonstration that this SMOTE-DP technique can produce synthetic data that not only ensures robust privacy protection but maintains utility in downstream learning tasks.
  arXiv  Detail & Related papers  (2025-06-02T17:27:10Z)
• Scaling Laws of Synthetic Data for Language Models [132.67350443447611]
  We introduce SynthLLM, a scalable framework that transforms pre-training corpora into diverse, high-quality synthetic datasets.<n>Our approach achieves this by automatically extracting and recombining high-level concepts across multiple documents using a graph algorithm.
  arXiv  Detail & Related papers  (2025-03-25T11:07:12Z)
• AugGen: Synthetic Augmentation Can Improve Discriminative Models [14.680260279598045]
  Synthetic data generation offers a promising alternative to external datasets or pre-trained models.<n>In this paper, we introduce AugGen, a self-contained synthetic augmentation technique.<n>Our findings demonstrate that carefully integrated synthetic data can both mitigate privacy constraints and substantially enhance discriminative performance in face recognition.
  arXiv  Detail & Related papers  (2025-03-14T16:10:21Z)
• Leveraging Programmatically Generated Synthetic Data for Differentially Private Diffusion Training [4.815212947276105]
  Programmatically generated synthetic data has been used in differential private training for classification to avoid privacy leakage.<n>The model trained with synthetic data generates unrealistic random images, raising challenges to adapt synthetic data for generative models.<n>We propose DPSynGen, which leverages generated synthetic data in diffusion models to address this challenge.
  arXiv  Detail & Related papers  (2024-12-13T04:22:23Z)
• Little Giants: Synthesizing High-Quality Embedding Data at Scale [71.352883755806]
  We introduce SPEED, a framework that aligns open-source small models to efficiently generate large-scale embedding data. SPEED uses only less than 1/10 of the GPT API calls, outperforming the state-of-the-art embedding model E5_mistral when both are trained solely on their synthetic data.
  arXiv  Detail & Related papers  (2024-10-24T10:47:30Z)
• Boosting Data Analytics With Synthetic Volume Expansion [3.568650932986342]
  This article explores the effectiveness of statistical methods on synthetic data and the privacy risks of synthetic data. A key finding within this framework is the generational effect, which reveals that the error rate of statistical methods on synthetic data decreases with the addition of more synthetic data but may eventually rise or stabilize.
  arXiv  Detail & Related papers  (2023-10-27T01:57:27Z)
• Reimagining Synthetic Tabular Data Generation through Data-Centric AI: A Comprehensive Benchmark [56.8042116967334]
  Synthetic data serves as an alternative in training machine learning models. ensuring that synthetic data mirrors the complex nuances of real-world data is a challenging task. This paper explores the potential of integrating data-centric AI techniques to guide the synthetic data generation process.
  arXiv  Detail & Related papers  (2023-10-25T20:32:02Z)
• Synthetic data, real errors: how (not) to publish and use synthetic data [86.65594304109567]
  We show how the generative process affects the downstream ML task. We introduce Deep Generative Ensemble (DGE) to approximate the posterior distribution over the generative process model parameters.
  arXiv  Detail & Related papers  (2023-05-16T07:30:29Z)
• Differentially Private Diffusion Models Generate Useful Synthetic Images [53.94025967603649]
  Recent studies have found that, by default, the outputs of some diffusion models do not preserve training data privacy. By privately fine-tuning ImageNet pre-trained diffusion models with more than 80M parameters, we obtain SOTA results on CIFAR-10 and Camelyon17. Our results demonstrate that diffusion models fine-tuned with differential privacy can produce useful and provably private synthetic data.
  arXiv  Detail & Related papers  (2023-02-27T15:02:04Z)
• Bias Mitigated Learning from Differentially Private Synthetic Data: A Cautionary Tale [13.881022208028751]
  Bias can affect all analyses as the synthetic data distribution is an inconsistent estimate of the real-data distribution. We propose several bias mitigation strategies using privatized likelihood ratios. We show that bias mitigation provides simple and effective privacy-compliant augmentation for general applications of synthetic data.
  arXiv  Detail & Related papers  (2021-08-24T19:56:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.