Private data sharing between decentralized users through the privGAN architecture

  title={Private data sharing between decentralized users through the privGAN architecture},
  author={Jean-François Rajotte and Raymond T. Ng},
  journal={2020 IEEE 24th International Enterprise Distributed Object Computing Workshop (EDOCW)},
  • J. Rajotte, R. Ng
  • Published 14 September 2020
  • Computer Science
  • 2020 IEEE 24th International Enterprise Distributed Object Computing Workshop (EDOCW)
More data is almost always beneficial for analysis and machine learning tasks. In many realistic situations however, an enterprise cannot share its data, either to keep a competitive advantage or to protect the privacy of the data sources, the enterprise’s clients for example. We propose a method for data owners to share synthetic or fake versions of their data without sharing the actual data, nor the parameters of models that have direct access to the data. The method proposed is based on the… 

Figures from this paper

Reducing bias and increasing utility by federated generative modeling of medical images using a centralized adversary

FELICIA (FEderated LearnIng with a Central Adversary) a generative mechanism enabling collaborative learning is introduced and it is shown how a data owner with limited and biased data could benefit from other data owners while keeping data from all the sources private.



PATE-GAN: Generating Synthetic Data with Differential Privacy Guarantees

This paper investigates a method for ensuring (differential) privacy of the generator of the Generative Adversarial Nets (GAN) framework, and modifies the Private Aggregation of Teacher Ensembles (PATE) framework and applies it to GANs.

Generative Models for Effective ML on Private, Decentralized Datasets

This paper demonstrates that generative models - trained using federated methods and with formal differential privacy guarantees - can be used effectively to debug many commonly occurring data issues even when the data cannot be directly inspected.

privGAN: Protecting GANs from membership inference attacks at low cost to utility

A novel GAN architecture that can generate synthetic data in a privacy preserving manner with minimal hyperparameter tuning and architecture selection is proposed and a theoretical understanding of the optimal solution of the privGAN loss function is provided.

Semi-supervised Knowledge Transfer for Deep Learning from Private Training Data

Private Aggregation of Teacher Ensembles (PATE) is demonstrated, in a black-box fashion, multiple models trained with disjoint datasets, such as records from different subsets of users, which achieves state-of-the-art privacy/utility trade-offs on MNIST and SVHN.

Differentially Private Federated Learning: A Client Level Perspective

The aim is to hide clients' contributions during training, balancing the trade-off between privacy loss and model performance, and empirical studies suggest that given a sufficiently large number of participating clients, this procedure can maintain client-level differential privacy at only a minor cost in model performance.

GS-WGAN: A Gradient-Sanitized Approach for Learning Differentially Private Generators

Gradient-sanitized Wasserstein Generative Adversarial Networks (GS-WGAN) is proposed, which allows releasing a sanitized form of the sensitive data with rigorous privacy guarantees, and is able to distort gradient information more precisely, and thereby enabling training deeper models which generate more informative samples.

Communication-Efficient Learning of Deep Networks from Decentralized Data

This work presents a practical method for the federated learning of deep networks based on iterative model averaging, and conducts an extensive empirical evaluation, considering five different model architectures and four datasets.

Differentially Private Generative Adversarial Network

This paper proposes a differentially private GAN (DPGAN) model, in which it is demonstrated that the method can generate high quality data points at a reasonable privacy level by adding carefully designed noise to gradients during the learning procedure.

LOGAN: Evaluating Privacy Leakage of Generative Models Using Generative Adversarial Networks

This paper presents the first membership inference attack on generative models, training a Generative Adversarial Network, which combines a discriminative and a generative model, to detect overfitting and recognize inputs that are part of training datasets by relying on the discriminator's capacity to learn statistical differences in distributions.

Private Post-GAN Boosting

Private post-GAN boosting (Private PGB), a differentially private method that combines samples produced by the sequence of generators obtained during GAN training to create a high-quality synthetic dataset, is proposed.