We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IT

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Theory

Title: PolyDot Coded Privacy Preserving Multi-Party Computation at the Edge

Abstract: We investigate the problem of privacy preserving distributed matrix multiplication in edge networks using multi-party computation (MPC). Coded multi-party computation (CMPC) is an emerging approach to reduce the required number of workers in MPC by employing coded computation. Existing CMPC approaches usually combine coded computation algorithms designed for efficient matrix multiplication with MPC. We show that this approach is not efficient. We design a novel CMPC algorithm; PolyDot coded MPC (PolyDot-CMPC) by using a recently proposed coded computation algorithm; PolyDot codes. We exploit "garbage terms" that naturally arise when polynomials are constructed in the design of PolyDot-CMPC to reduce the number of workers needed for privacy-preserving computation. We show that entangled polynomial codes, which are consistently better than PolyDot codes in coded computation setup, are not necessarily better than PolyDot-CMPC in MPC setting.
Comments: 23 pages, 5 figures
Subjects: Information Theory (cs.IT); Distributed, Parallel, and Cluster Computing (cs.DC)
Cite as: arXiv:2106.08290 [cs.IT]
  (or arXiv:2106.08290v4 [cs.IT] for this version)

Submission history

From: Elahe Vedadi [view email]
[v1] Tue, 15 Jun 2021 17:04:03 GMT (1141kb)
[v2] Mon, 12 Jul 2021 19:01:28 GMT (1139kb)
[v3] Tue, 20 Jul 2021 21:52:19 GMT (1139kb)
[v4] Tue, 15 Mar 2022 02:44:19 GMT (768kb,D)

Link back to: arXiv, form interface, contact.