Privacy-Preserving Machine Learning

We consider the problem of privately learning a sparse model across multiple sensitive datasets, and propose learning individual models locally and privately aggregating them using secure multi-party computation. Our project combines ideas from distributed machine learning with secure multi-party computation.

Papers

Lu Tian, Bargav Jayaraman, Quanquan Gu, and David Evans. Aggregating Private Sparse Learning Models Using Multi-Party Computation. In Private Multi‑Party Machine Learning (NIPS 2016 Workshop), Barcelona, 9 December 2016. (PDF, 6 pages)

Code

(Coming Soon - please contact us if you would like early access.)