DTGAN: Differential Private Training for Tabular GANs

Figure 1: Privacy Preserving Generator

1. Introduction

2. DTGAN

  • The discriminator directly interacts with real data.
  • The discriminator gradient norms are directly bounded due to the gradient penalty.
  • Subsampling is highly efficient as it is defined as 𝛾=𝐵/𝑁 where 𝐵 is the batch size and 𝑁 is the size of the training dataset.
  • The use of the Wasserstein loss requires multiple updates to the discriminator for performing a single update to the generator.
  • Training multiple discriminators to perform distributed GAN training increases the privacy cost significantly.
  • DP-generator makes for a safe public release after training.
  • Distributed discriminators do not increase privacy cost.
  • Subsampling introduces complexity via multiple discriminators and is defined as 1/N_d where N_d is the number of discriminators.
  • Added loss functions on the generator increase privacy cost.

3. Inference Attacks

Figure 2: Membership Inference Attack
Figure 3: Attribute Inference Attack

4. Results

Table 1: Difference of accuracy (%), F1-score, AUC and APR between original and synthetic data: average over 3 different datasets and different privacy budgets, epsilon = 1 & 100.
Table 2: Statistical similarity metrics averaged on 3 datasets with different privacy budget, epsilon = 1 & 100.
Table 3: Empirical privacy gain against membership attack with naïve and correlation feature extraction, and attribute inference attack: average over 3 different datasets with privacy budget, epsilon = 1

5. Conclusion

--

--

--

I am a researcher at Generatrix- An AI-based privacy preserving data synthesizing platform. I have an avid passion for new and emerging technologies in AI & ML

Love podcasts or audiobooks? Learn on the go with our new app.

Recommended from Medium

Federated Learning with Secure Aggregation in TensorFlow

Reinforcement Learning:With Q Learning Implementation in Python

Event Wrap-up: WTF is… Machine Learning?

Protein Folding and P vs NP (a classic)

Easy Explanation on Unsupervised Learning

Learning Day 20: Batch normalization concept and usage in Pytorch

Neural Gas Simplified

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store
Aditya Kunar

Aditya Kunar

I am a researcher at Generatrix- An AI-based privacy preserving data synthesizing platform. I have an avid passion for new and emerging technologies in AI & ML

More from Medium

How to enable TensorFlow to use GPU drivers on Windows

Image Inpaiting — Next GEN Watermark needed?

IoT And Measurement Devices

get upper and lower limit from imcontrast