Helen: Maliciously Secure Cooperative Learning for Linear Models

C4DT – IC Talk

By Raluca Ada Popa, Assistant Professor of Computer Science at UC Berkeley


Many organizations wish to collaboratively train machine learning models on their combined datasets for a common benefit (e.g., better medical research, or fraud detection). However, they often cannot share their plaintext datasets due to privacy concerns and/or business competition. In this talk, Helen will be presented, a system that allows multiple parties to train a linear model without revealing their data, a setting we call coopetitive learning. Compared to prior secure training systems, Helen protects against a much stronger adversary who is malicious and can compromise m − 1 out of m parties.

Thursday June 20th, 2019 @10:15 room BC 420 (see map)