English

Sparsity and Privacy in Secret Sharing: A Fundamental Trade-Off

Cryptography and Security 2023-08-15 v1 Distributed, Parallel, and Cluster Computing Information Theory math.IT

Abstract

This work investigates the design of sparse secret sharing schemes that encode a sparse private matrix into sparse shares. This investigation is motivated by distributed computing, where the multiplication of sparse and private matrices is moved from a computationally weak main node to untrusted worker machines. Classical secret-sharing schemes produce dense shares. However, sparsity can help speed up the computation. We show that, for matrices with i.i.d. entries, sparsity in the shares comes at a fundamental cost of weaker privacy. We derive a fundamental tradeoff between sparsity and privacy and construct optimal sparse secret sharing schemes that produce shares that leak the minimum amount of information for a desired sparsity of the shares. We apply our schemes to distributed sparse and private matrix multiplication schemes with no colluding workers while tolerating stragglers. For the setting of two non-communicating clusters of workers, we design a sparse one-time pad so that no private information is leaked to a cluster of untrusted and colluding workers, and the shares with bounded but non-zero leakage are assigned to a cluster of partially trusted workers. We conclude by discussing the necessity of using permutations for matrices with correlated entries.

Keywords

Cite

@article{arxiv.2308.06413,
  title  = {Sparsity and Privacy in Secret Sharing: A Fundamental Trade-Off},
  author = {Rawad Bitar and Maximilian Egger and Antonia Wachter-Zeh and Marvin Xhemrishi},
  journal= {arXiv preprint arXiv:2308.06413},
  year   = {2023}
}