The experiment successfully developed a federated deep learning solution, enabling multiple Digital Innovation Hubs (DIHs) to collaborate on an AI tool. This was achieved by locally training neural networks on their data and then amalgamating these into a larger, more efficient solution. Initially, the solution was created as a proof of concept using an example dataset and deep learning model. This model was designed with the flexibility to be updated for more applicable use cases, particularly those relevant to the manufacturing sector.

Furthermore, the experiment explored the feasibility of establishing this kind of federated deep learning solution as a reusable component for DIHs. The solution was built based on the International Data Spaces (IDS) and incorporated a dataspace for DIHs, planned for deployment in a later stage of the experimentation. Additionally, the DIHs connected to this dataspace were equipped to deploy their own connectors to local partners, thus facilitating broader participation and integration into the platform. This approach not only demonstrated the practicality of federated deep learning in a collaborative environment but also set a precedent for future applications and innovations in the field.