On the Upload versus Download Cost for Secure and Private Matrix Multiplication

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 , with a matrix , where . The set of candidate matrices are public, and available at all the servers. The goal of the user is to distributedly compute , such that no information is leaked about the matrix to any server; and the index 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: for 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.
View on arXiv