Feb. 6, 2024, 5:11 a.m. | Monika Henzinger Jalaj Upadhyay Sarvagya Upadhyay

cs.CR updates on arXiv.org arxiv.org

The first large-scale deployment of private federated learning uses differentially private counting in the continual release model as a subroutine (Google AI blog titled "Federated Learning with Formal Differential Privacy Guarantees"). In this case, a concrete bound on the error is very relevant to reduce the privacy parameter. The standard mechanism for continual counting is the binary mechanism. We present a novel mechanism and show that its mean squared error is both asymptotically optimal and a factor 10 smaller than …

blog case concrete cs.cr cs.ds cs.lg deployment differential privacy error federated federated learning google google ai large mechanism parameter privacy private release relevant scale standard

Information Technology Specialist I, LACERA: Information Security Engineer

@ Los Angeles County Employees Retirement Association (LACERA) | Pasadena, CA

Issues Management & Risk Treatment Sr. Consultant

@ Northern Trust | Tempe, AZ Building 2190

Dir. DDIT ISC Enterprise Architecture AppSec

@ Novartis | Hyderabad (Office)

System Access Management Manager

@ Ingram Micro | CA - Irvine, HQ

Oracle Linux Systems Administrator

@ Leidos | 1662 Intelligence Community Campus - Bethesda MD

Senior Systems Engineer - AWS

@ CACI International Inc | 999 REMOTE