Acs, G;
Melis, L;
Castelluccia, C;
De Cristofaro, E;
(2018)
Differentially Private Mixture of Generative Neural Networks.
IEEE Transactions on Knowledge and Data Engineering
, 31
(6)
pp. 1109-1121.
10.1109/TKDE.2018.2855136.
Preview |
Text
De Cristofaro_Differentially Private Mixture of Generative Neural Networks_AAM.pdf - Accepted Version Download (855kB) | Preview |
Abstract
Generative models are used in a wide range of applications building on large amounts of contextually rich information. Due to possible privacy violations of the individuals whose data is used to train these models, however, publishing or sharing generative models is not always viable. In this paper, we present a novel technique for privately releasing generative models and entire high-dimensional datasets produced by these models. We model the generator distribution of the training data with a mixture of k generative neural networks. These are trained together and collectively learn the generator distribution of a dataset. Data is divided into k clusters, using a novel differentially private kernel k-means, then each cluster is given to separate generative neural networks, such as Restricted Boltzmann Machines or Variational Autoencoders, which are trained only on their own cluster using differentially private gradient descent. We evaluate our approach using the MNIST dataset, as well as call detail records and transit datasets, showing that it produces realistic synthetic samples, which can also be used to accurately compute arbitrary number of counting queries.
| Type: | Article |
|---|---|
| Title: | Differentially Private Mixture of Generative Neural Networks |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1109/TKDE.2018.2855136 |
| Publisher version: | https://doi.org/10.1109/TKDE.2018.2855136 |
| Language: | English |
| Additional information: | This version is the author accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Data models, Privacy, Data privacy, Training, Neural networks, Kernel, Machine learning |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10052339 |
Loading...
Loading...Loading...Loading...Archive Staff Only
 |
View Item |
