With our host lab (the Machine Learning and Optimization group) we contribute to the DISCO platform, an open-access, open source  Distributed Collaborative learning platform, offering users a means to collaboratively build machine learning models without sharing any original data. The modular design of DISCO allows users to leverage a flexible combination of features according to their requirements. Such as offering choices on federated and decentralised communication protocols, various levels of privacy guarantees and several approaches to weight aggregation strategies that allow model personalization
and resilience to unseen bias in the federation