Secure Two-Party Feature Selection
Cryptography and Security
2019-01-04 v1
Abstract
In this work, we study how to securely evaluate the value of trading data without requiring a trusted third party. We focus on the important machine learning task of classification. This leads us to propose a provably secure four-round protocol that computes the value of the data to be traded without revealing the data to the potential acquirer. The theoretical results demonstrate a number of important properties of the proposed protocol. In particular, we prove the security of the proposed protocol in the honest-but-curious adversary model.
Cite
@article{arxiv.1901.00832,
title = {Secure Two-Party Feature Selection},
author = {Vanishree Rao and Yunhui Long and Hoda Eldardiry and Shantanu Rane and Ryan Rossi and Frank Torres},
journal= {arXiv preprint arXiv:1901.00832},
year = {2019}
}