Big data forms the fundamental basis for the success of Machine Learning. Yet, a large amount of the world’s digitized data is locked up in data silos, leaving its potential untapped. Federated Machine Learning is a novel Machine Learning paradigm with the potential to overcome data silos by enabling the decentralized training of Machine Learning models through a model-to-data approach. Despite its potential advantages, most Federated Machine Learning projects fail to actualize due to their decentralized structure and incomprehensive interrelations. Current literature lacks clear guidelines on which steps need to be performed to successfully implement Federated Machine Learning projects. This study aims to close this research gap. Through a design science research approach, we provide three distinct activity models which outline required tasks in the development of Federated Machine Learning systems. Thereby, we aim to reduce complexity and ease the implementation process by guiding practitioners through the project life cycle.
Müller, Tobias; Zahn, Milena; and Matthes, Florian, "A Pathway for the Practical Adoption of Federated Machine Learning Projects" (2023). PACIS 2023 Proceedings. 6.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.