Differential privacy based on importance weighting

By Zhanglong Ji, Charles Peter Elkan. 2013

43 citations recorded by the source corpus.

This paper analyzes a novel method for publishing data while still protecting privacy. The method is based on computing weights that make an existing dataset, for which there are no confidentiality issues, analogous to the dataset that must be kept private. The existing dataset may be genuine but public already, or it may be synthetic. The weights are importance sampling weights, but to protect privacy, they are regularized and have noise added. The weights allow statistical queries to be answered approximately while provably guaranteeing differential privacy. We derive an expression for the asymptotic variance of the approximate answers. Experiments show that the new mechanism performs well even when the privacy budget is small, and when the public and private datasets are drawn from different populations.

Subfields: Artificial Intelligence, Sociology and Political Science

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Sociology and Political Science

Citations in the corpus

The Power of the Hybrid Model for Mean Estimation (2018, neutral): “[36].”
Differentially Private Hypothesis Transfer Learning (2018, methodology): “To better incorporate the source hypotheses, we also adapt an importance weighting mechanism in [16] to measure the relationship between sources and target by leveraging a public auxiliary unlabeled data set.”
Differential Privacy and Machine Learning: a Survey and Review (2014, methodology): “For example, [52, 40, 46, 28, 30] use the Laplacian mechanism, while”
Differentially Private Fine-tuning of Language Models (2021, background): “Finally, other works (both theoretical and experimental) investigate the role of public data in private query release, synthetic data generation, and prediction (Ji & Elkan, 2013; Beimel et al., 2016; Alon et al., 2019; Nandi & Bassily, 2020; Bassily et al., 2020a;b; Liu et al., 2021).”