0000000000242054
AUTHOR
Derian Boer
Secure Sum Outperforms Homomorphic Encryption in (Current) Collaborative Deep Learning
Deep learning (DL) approaches are achieving extraordinary results in a wide range of domains, but often require a massive collection of private data. Hence, methods for training neural networks on the joint data of different data owners, that keep each party's input confidential, are called for. We address a specific setting in federated learning, namely that of deep learning from horizontally distributed data with a limited number of parties, where their vulnerable intermediate results have to be processed in a privacy-preserving manner. This setting can be found in medical and healthcare as well as industrial applications. The predominant scheme for this is based on homomorphic encryption…
Privacy Preserving Client/Vertical-Servers Classification
We present a novel client/vertical-servers architecture for hybrid multi-party classification problem. The model consists of clients whose attributes are distributed on multiple servers and remain secret during training and testing. Our solution builds privacy-preserving random forests and completes them with a special private set intersection protocol that provides a central commodity server with anonymous conditional statistics. Subsequently, the private set intersection protocol can be used to privately classify the queries of new clients using the commodity server’s statistics. The proviso is that the commodity server must not collude with other parties. In cases where this restriction …