On Differentially Private Linear Algebra

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We introduce efficient differentially private (DP) algorithms for several linear algebraic tasks, including solving linear equalities over arbitrary fields, linear inequalities over the reals, and computing affine spans and convex hulls. As an application, we obtain efficient DP algorithms for learning halfspaces and affine subspaces. Our algorithms addressing equalities are strongly polynomial, whereas those addressing inequalities are weakly polynomial. Furthermore, this distinction is inevitable: no DP algorithm for linear programming can be strongly polynomial-time efficient.

Original languageEnglish
Title of host publicationSTOC 2025 - Proceedings of the 57th Annual ACM Symposium on Theory of Computing
EditorsMichal Koucky, Nikhil Bansal
PublisherAssociation for Computing Machinery
Pages2362-2373
Number of pages12
ISBN (Electronic)9798400715105
DOIs
StatePublished - 15 Jun 2025
Event57th Annual ACM Symposium on Theory of Computing, STOC 2025 - Prague, Czech Republic
Duration: 23 Jun 202527 Jun 2025

Publication series

NameProceedings of the Annual ACM Symposium on Theory of Computing
ISSN (Print)0737-8017

Conference

Conference57th Annual ACM Symposium on Theory of Computing, STOC 2025
Country/TerritoryCzech Republic
CityPrague
Period23/06/2527/06/25

Keywords

  • differential privacy
  • linear algebra

Fingerprint

Dive into the research topics of 'On Differentially Private Linear Algebra'. Together they form a unique fingerprint.

Cite this