|
Activity Number:
|
85
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 31, 2016 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #319367
|
View Presentation
|
|
Title:
|
ReASSESS: A Robust Adaptive Mechanism Using Subsetting and Multiplicative Weights in Data Synthesis
|
|
Author(s):
|
Evercita Cuevas Eugenio* and Fang Liu
|
|
Companies:
|
University of Notre Dame and University of Notre Dame
|
|
Keywords:
|
differential privacy ;
adaptive mechanism ;
statistical disclosure limitation ;
data synthesis ;
multiplicative weights
|
|
Abstract:
|
Balancing between protecting the privacy of individuals who contribute to datasets and releasing datasets of good utility can be difficult. Concern exists that an intruder may identify a subject in a released data set. Differential privacy provides a conceptual approach to bring strong mathematical guarantee for privacy protection. Currently available differentially private data synthesis methods often independently sanitize queries via the Laplace and exponential mechanisms to generate synthetic data with differential privacy ensured. Since the correlation between queries is unaccounted for, excessive noise may be added to the released results. In this talk, we will propose a novel nonparametric adaptive mechanism that uses multiplicative weights and subsetting to generate a distribution that is similar to the distribution of the original data by certain standards. This adaptive mechanism considers the relationships between queries to preserve as much of the original information as possible when releasing synthetic data while still ensuring differential privacy.
|
Authors who are presenting talks have a * after their name.
