Bicoptor: Two-round Secure Three-party Non-linear Computation without Preprocessing for Privacy-preserving Machine Learning

arXiv:2210.01988v3 Announce Type: replace
Abstract: The overhead of non-linear functions dominates the performance of the secure multiparty computation (MPC) based privacy-preserving machine learning (PPML). This work introduces a family of novel secure three-party computation (3PC) protocols, Bicoptor, which improve the efficiency of evaluating non-linear functions. The basis of Bicoptor is a new sign determination protocol, which relies on a clever use of the truncation protocol proposed in SecureML (S\&P 2017). Our 3PC sign determination protocol only requires two communication rounds, …

arxiv computation cs.cr efficiency family functions linear machine machine learning mpc non novel party performance privacy protocols secure multiparty computation work