Yupeng Zhang

TL;DR: This paper presents new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method, and implements the first privacy preserving system for training neural networks.

TL;DR: In this article, the authors present new and efficient protocols for privacy preserving machine learning for linear regression, logistic regression and neural network training using the stochastic gradient descent method, where data owners distribute their private data among two non-colluding servers who train various models on the joint data using secure two-party computation.

TL;DR: In this paper, file-injection attacks on the query privacy of searchable encryption (SE) schemes have been studied, in which the server sends files to the client that the client then encrypts and stores, and such attacks can reveal the client queries in their entirety using very few injected files.

TL;DR: In this article, the authors present vSQL, a cryptographic protocol for publicly verifiable SQL queries on dynamic databases, which relies on two extensions of the CMT interactive-proof protocol: (i) supporting outsourced input via the use of a polynomial-delegation protocol with succinct proofs, and (ii) supporting auxiliary input efficiently.

TL;DR: Libra is presented, the first zero-knowledge proof system that has both optimal prover time and succinct proof size/verification time and an one-time trusted setup that depends only on the size of the input to the circuit and not on the circuit logic.