The AI4EOSC project was showcased at the world’s largest conference on Federated Learning (FL), the Flower AI Summit 2024, held in London on March 14-15.
The conference featured not only the advancements within Federated Learning but also the collaborative efforts driving innovation across the AI landscape and different applications.
On the first day, dedicated to AI research, we introduced the European Open Science Cloud (EOSC) ecosystem, presented the AI4EOSC project, and talked about how we have implemented Federated Learning in the AI4EOSC platform, whose first release, AI4EOSC 1, took place earlier this month. During her talk (“Federated AI in the European Open Science Cloud”) our colleague Judith Sáinz-Pardo Díaz (CSIC) seized the opportunity to present some modifications that have been made from AI4EOSC to the Flower library in order to allow client authentication with the FL server and a secret management system for this purpose, in order to add an additional layer of privacy.
The second day was dedicated to AI in the industry, with the participation of top level companies, exposing their uses and applications of FL, such as AWS, NVIDIA, Samsung AI, etc
Judith Sáinz-Pardo (CSIC) during her presentation at the Flower summit.
This two-day event has positioned AI4EOSC as a platform for the development of AI/ML/DL models with an enhanced set of tools for model development, monitoring and inference, but with a special focus on the use and implementation of Federated Learning. With the use of FL, AI4EOSC aims to provide users with the possibility of training models without centralizing the data from the different data owners involved in creating a global model, thus promoting privacy preserving ML/DL applications.
Particularly notable is Flower’s role within the AI4EOSC project, where this library is instrumental in enabling users to conduct Federated Learning training effortlessly and intuitively. Through AI4EOSC, users can launch a Flower server without dealing with code and configuration details, just directly from the web dashboard interface. Once the server is deployed within AI4EOSC, the different clients (data owners) can run their code from other deployments on the platform, locally, or on other cloud providers. This streamlined process empowers users to leverage Federated Learning capabilities with ease and flexibility, further advancing the frontier of privacy preserving ML/DL technologies.
Flower is a Python framework designed for training AI/ML/DL models using a Federated Learning architecture. Flower’s primary goal is to offer users a unified approach to FL, analytics, and evaluation.
