A1309
Title: Federated learning of quantile inference under local differential privacy
Authors: Shuyuan Wu - Shanghai University of Finance and Economics (China) [presenting]
Leheng Cai - Tsinghua University (China)
Qirui Hu - Tsinghua University (China)
Abstract: Federated learning for quantile inference under local differential privacy (LDP) is investigated. An estimator based on local stochastic gradient descent (SGD) is proposed, whose local gradients are perturbed via a randomized mechanism with global parameters, making the procedure tolerant of communication and storage constraints without compromising statistical efficiency. Although the quantile loss and its corresponding gradient do not satisfy standard smoothness conditions typically assumed in existing literature, asymptotic normality for the estimator as well as a functional central limit theorem are established. The proposed method accommodates data heterogeneity and allows each server to operate with an individual privacy budget. Furthermore, confidence intervals for the target value are constructed through a selfnormalization approach, thereby circumventing the need to estimate additional nuisance parameters. Extensive numerical experiments and real data application validate the theoretical guarantees of the proposed methodology.
