A Knowledge Transfer Framework for Differentially Private Sparse Learning
This work addresses privacy-preserving machine learning for high-dimensional data, offering incremental improvements in utility for sparse learning tasks.
The paper tackles the problem of estimating high-dimensional sparse models while preserving training data privacy, achieving improved utility guarantees for sparse linear and logistic regression compared to prior results.
We study the problem of estimating high dimensional models with underlying sparse structures while preserving the privacy of each training example. We develop a differentially private high-dimensional sparse learning framework using the idea of knowledge transfer. More specifically, we propose to distill the knowledge from a "teacher" estimator trained on a private dataset, by creating a new dataset from auxiliary features, and then train a differentially private "student" estimator using this new dataset. In addition, we establish the linear convergence rate as well as the utility guarantee for our proposed method. For sparse linear regression and sparse logistic regression, our method achieves improved utility guarantees compared with the best known results (Kifer et al., 2012; Wang and Gu, 2019). We further demonstrate the superiority of our framework through both synthetic and real-world data experiments.