Exploiting Unintended Feature Leakage in Collaborative Learning
IEEE Symposium on Security and Privacy, 2019.
EI
Abstract:
Collaborative machine learning and related techniques such as federated learning allow multiple participants, each with his own training dataset, to build a joint model by training locally and periodically exchanging model updates.We demonstrate that these updates leak unintended information about participantsu0027 training data and devel...More
Code:
Data:
Tags
Comments