Timezone: »

 
Distributional Privacy for Data Sharing
Zinan Lin · Shuaiqi Wang · Vyas Sekar · Giulia Fanti

Fri Dec 02 07:48 AM -- 07:50 AM (PST) @
Event URL: https://openreview.net/forum?id=6oVAzFsHLFK »

Data sharing between different parties has become an important engine powering modern research and development processes. An important class of privacy concerns in data sharing regards the underlying distribution of data. For example, the total traffic volume of data from a networking company reveals the scale of its business. Unfortunately, existing privacy frameworks do not adequately address this class of concerns. In this paper, we propose distributional privacy, a framework for analyzing and protecting these distributional privacy concerns in data sharing scenarios. Distributional privacy is applicable in multiple data sharing settings, including synthetic data release. Theoretically, we analyze the lower and upper bounds of privacy-distortion trade-offs. Practically, we propose data release mechanism for protecting distributional privacy concerns, and demonstrate that they achieve better privacy-distortion trade-offs than alternative privacy mechanisms on real-world datasets.

Author Information

Zinan Lin (Microsoft Research, Carnegie Mellon University)
Shuaiqi Wang (CMU, Carnegie Mellon University)

Shuaiqi Wang is a Ph.D. student in the Electrical and Computer Engineering Department at Carnegie Mellon University, under the supervision of Giulia Fanti. Prior to joining Carnegie Mellon University in 2021, he received the BE degree from the department of Computer Science at Shanghai Jiao Tong University in 2020. Shuaiqi’s research interest is in the theoretical foundations of machine learning, and the applications in privacy, security, federated learning, and data sharing. He was awarded the Carnegie Institute of Technology Dean’s Fellow in 2021.

Vyas Sekar (Carnegie Mellon University)
Giulia Fanti (CMU)

More from the Same Authors