2026-07-22T04:54:18Z https://drops.dagstuhl.de/oai

oai:drops-oai.dagstuhl.de:16530 2024-03-06T10:56:14Z ddc:004 open_access

Differential Secrecy for Distributed Data and Applications to Robust Differentially Secure Vector Summation Talwar, Kunal Zero Knowledge Secure Summation Differential Privacy Computing the noisy sum of real-valued vectors is an important primitive in differentially private learning and statistics. In private federated learning applications, these vectors are held by client devices, leading to a distributed summation problem. Standard Secure Multiparty Computation protocols for this problem are susceptible to poisoning attacks, where a client may have a large influence on the sum, without being detected. In this work, we propose a poisoning-robust private summation protocol in the multiple-server setting, recently studied in PRIO [Henry Corrigan-Gibbs and Dan Boneh, 2017]. We present a protocol for vector summation that verifies that the Euclidean norm of each contribution is approximately bounded. We show that by relaxing the security constraint in SMC to a differential privacy like guarantee, one can improve over PRIO in terms of communication requirements as well as the client-side computation. Unlike SMC algorithms that inevitably cast integers to elements of a large finite field, our algorithms work over integers/reals, which may allow for additional efficiencies. Schloss Dagstuhl – Leibniz-Zentrum für Informatik Kunal Talwar 2022 Is Part Of LIPIcs, Volume 218, 3rd Symposium on Foundations of Responsible Computing (FORC 2022) InProceedings Text doc-type:ResearchArticle publishedVersion application/pdf doi:10.4230/LIPIcs.FORC.2022.7 urn:nbn:de:0030-drops-165302 https://drops.dagstuhl.de/entities/document/10.4230/LIPIcs.FORC.2022.7 eng https://creativecommons.org/licenses/by/4.0/legalcode