Federated machine learning through edge ready architectures with privacy preservation as a service

Abstract

This paper presents the details of a novel approach, based on edge and advanced privacy preserving solutions, that tries to accelerate the adoption of personal data federation for the benefit of the evolution of valuable advanced AI models. The approach focuses on the establishment of high degree of trust between data owner and data management infrastructure so that consent in data processing is given by means of functional and enforceable options applicable at all levels of workloads and processes. The overall set of solutions will be delivered as an open-source set of implementations in the context of the PAROMA-MED project.