TY - GEN
T1 - On the Upload versus Download Cost for Secure and Private Matrix Multiplication
AU - Chang, Wei Ting
AU - Tandon, Ravi
N1 - Funding Information:
This work was supported by NSF Grants CAREER 1651492, and CNS 1715947.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/8
Y1 - 2019/8
N2 - In this paper, we study the problem of secure and private distributed matrix multiplication. Specifically, we focus on a scenario where a user wants to compute the product of a confidential matrix A, with a matrix Bθ, where θ 1,..., M. The set of candidate matrices B1,..., BM are public, and available at all the N servers. The goal of the user is to distributedly compute ABθ, such that (a) no information is leaked about the matrix A to any server; and (b) the index θ is kept private from each server. Our goal is to understand the fundamental tradeoff between the upload vs download cost for this problem. Our main contribution is to show that the lower convex hull of following (upload, download) pairs: (U,D) = (N/(K-1), (K/(K-1)) (1 + (K/N) + ··· + (K/N)M-1)) for K = 2,..., N is achievable. The scheme improves upon state-of-the-art existing schemes for this problem, and leverages ideas from secret sharing and coded private information retrieval.
AB - In this paper, we study the problem of secure and private distributed matrix multiplication. Specifically, we focus on a scenario where a user wants to compute the product of a confidential matrix A, with a matrix Bθ, where θ 1,..., M. The set of candidate matrices B1,..., BM are public, and available at all the N servers. The goal of the user is to distributedly compute ABθ, such that (a) no information is leaked about the matrix A to any server; and (b) the index θ is kept private from each server. Our goal is to understand the fundamental tradeoff between the upload vs download cost for this problem. Our main contribution is to show that the lower convex hull of following (upload, download) pairs: (U,D) = (N/(K-1), (K/(K-1)) (1 + (K/N) + ··· + (K/N)M-1)) for K = 2,..., N is achievable. The scheme improves upon state-of-the-art existing schemes for this problem, and leverages ideas from secret sharing and coded private information retrieval.
KW - Distributed Matrix Multiplication
KW - Private Information Retrieval
KW - Secure Matrix Multiplication
UR - http://www.scopus.com/inward/record.url?scp=85081053857&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85081053857&partnerID=8YFLogxK
U2 - 10.1109/ITW44776.2019.8989342
DO - 10.1109/ITW44776.2019.8989342
M3 - Conference contribution
AN - SCOPUS:85081053857
T3 - 2019 IEEE Information Theory Workshop, ITW 2019
BT - 2019 IEEE Information Theory Workshop, ITW 2019
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2019 IEEE Information Theory Workshop, ITW 2019
Y2 - 25 August 2019 through 28 August 2019
ER -