Scalable and secure distributed computation networks

Purpose and goal:

The aim of this project is to develop secure and scalable distributed computation networks based on AI-in-a-box computation nodes for secure, privacy-preserving, and scalable machine learning on open infrastructure. This is achieved by leveraging a combination of homomorphic encryption, differential privacy, as well as federated learning.

Expected results and effects:

This project develops competence within development and implementation of secure computation networks for secure machine learning at scale, including the required machine learning methods, as well as an overview of the data security requirements for processing of sensitive data on open infrastructure.

Approach and implementation: The project´s duration is three years and consists of the following work packages: * WP1: Compuation network - development of the computation network architecture and its nodes * WP2: Algorithms - algorithm research and development * WP3: Application platform - design and development of the user application

* WP4: Use-case - implementation and evaluation of the algorithms, architecture, and network in a proof-of-concept

In addition to the technical work packages above, a fifth work package is dedicated to project management, reporting, and dissemination.