Mean Estimation with User-level Privacy under Data Heterogeneity
This addresses privacy-preserving mean estimation for applications like language data with diverse user contributions, but it is incremental as it builds on existing differential privacy frameworks.
The paper tackles the problem of estimating population-level mean from heterogeneous user data, where users differ in both data distribution and quantity, while preserving user-level differential privacy, and demonstrates asymptotic optimality of the estimator with proven lower bounds on error.
A key challenge in many modern data analysis tasks is that user data are heterogeneous. Different users may possess vastly different numbers of data points. More importantly, it cannot be assumed that all users sample from the same underlying distribution. This is true, for example in language data, where different speech styles result in data heterogeneity. In this work we propose a simple model of heterogeneous user data that allows user data to differ in both distribution and quantity of data, and provide a method for estimating the population-level mean while preserving user-level differential privacy. We demonstrate asymptotic optimality of our estimator and also prove general lower bounds on the error achievable in the setting we introduce.