|
Activity Number:
|
483
- Privacy and Work Force
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 10, 2022 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #323059
|
|
|
Title:
|
Missing Data Imputation Under Differential Privacy
|
|
Author(s):
|
Soumojit Das* and Shawn Merrill
|
|
Companies:
|
University of Maryland, College Park and University of Maryland, College Park
|
|
Keywords:
|
Differential Privacy;
Missing Data;
Imputation
|
|
Abstract:
|
Missing data is a common problem in many existing data products, particularly in sample surveys. Solutions generally rely on incorporating the available data in ways to make inferences in the presence of missing values. This heavier reliance on the existing data, and the incorporation of the existing data in models used to infer the missing data, causes problems with differential privacy and the need to protect the impact of any individual as some individuals now have out-sized influence. Our work focuses on making differentially private inferences from data while imputing these missing values.
|
Authors who are presenting talks have a * after their name.
