Data Science Seminar Federated Learning - An Introduction and Sample Applications

Federated Learning allows decentralized model training by keeping data on local devices, addressing challenges like regulatory restrictions and bandwidth limits that hinder centralized data sharing. Clients improve models using their local data and send encrypted updates to a central server, where these are combined into a global model. This approach is widely used in applications such as personalized healthcare, enhancing fraud detection and credit scoring models while maintaining user privacy, and optimizing mobile device performance (e.g., keyboard prediction) through collaborative learning. This introductory video on Federated Learning is part of our customizable one- to two-day workshop, featuring tailored presentations, discussions, and practical exercises based on your use cases. We're happy to collaborate to refine or identify the best use cases for your organization. For more information, please visit: https://www.itwm.fraunhofer.de/en/departments/fm/latest-news/workshop-federated-learning-en.html