Differentially Private Empirical Risk Minimization with Sparsity-Inducing Norms
This work addresses privacy-preserving machine learning for data with structured sparsity, but it is incremental as it extends existing differential privacy methods to new regularizers and dual problems.
The paper tackles differentially private empirical risk minimization with sparsity-inducing norms by analyzing algorithms like output perturbation and Frank-Wolfe, deriving excess risk bounds that depend on the Gaussian width of the dual norm's unit ball.
Differential privacy is concerned about the prediction quality while measuring the privacy impact on individuals whose information is contained in the data. We consider differentially private risk minimization problems with regularizers that induce structured sparsity. These regularizers are known to be convex but they are often non-differentiable. We analyze the standard differentially private algorithms, such as output perturbation, Frank-Wolfe and objective perturbation. Output perturbation is a differentially private algorithm that is known to perform well for minimizing risks that are strongly convex. Previous works have derived excess risk bounds that are independent of the dimensionality. In this paper, we assume a particular class of convex but non-smooth regularizers that induce structured sparsity and loss functions for generalized linear models. We also consider differentially private Frank-Wolfe algorithms to optimize the dual of the risk minimization problem. We derive excess risk bounds for both these algorithms. Both the bounds depend on the Gaussian width of the unit ball of the dual norm. We also show that objective perturbation of the risk minimization problems is equivalent to the output perturbation of a dual optimization problem. This is the first work that analyzes the dual optimization problems of risk minimization problems in the context of differential privacy.