Paper 2019/1246

A Note on Our Submission to Track 4 of iDASH 2019

Marcel Keller and Ke Sun

Abstract

iDASH is a competition soliciting implementations of cryptographic schemes of interest in the context of biology. In 2019, one track asked for multi-party computation implementations of training of a machine learning model suitable for two datasets from cancer research. In this note, we describe our solution submitted to the competition. We found that the training can be run on three AWS c5.9xlarge instances in less then one minute using MPC tolerating one semi-honest corruption, and less than ten seconds at a slightly lower accuracy. After seeing some winning solutions, we have lowered this figure to less than a second.

Note: Improved results using approximate sigmoid function

Metadata
Available format(s)
PDF
Category
Implementation
Publication info
Preprint. MINOR revision.
Keywords
Machine learningmulti-party computation
Contact author(s)
marcel keller @ data61 csiro au
History
2020-03-30: revised
2019-10-24: received
See all versions
Short URL
https://ia.cr/2019/1246
License
Creative Commons Attribution
CC BY

BibTeX

@misc{cryptoeprint:2019/1246,
      author = {Marcel Keller and Ke Sun},
      title = {A Note on Our Submission to Track 4 of iDASH 2019},
      howpublished = {Cryptology ePrint Archive, Paper 2019/1246},
      year = {2019},
      note = {\url{https://eprint.iacr.org/2019/1246}},
      url = {https://eprint.iacr.org/2019/1246}
}
Note: In order to protect the privacy of readers, eprint.iacr.org does not use cookies or embedded third party content.