Data Science Seminar: Federated Learning: The future of Edge Intelligence is now!

View comments (1)


In recent years, the ever-increasing resource capacities and allocated data at the network's edge encouraged shifting the data analytics to new generations of big data decentralised systems. Aiming at enabling on-device collaborative training of distributed machine learning and artificial intelligence models, edge intelligence came to realise. Federated Learning (FL) emerged as a popular privacy-preserving approach across this realm, aiming at conducting inference when data are decentralised and locally stored on several distributed nodes, under two main constraints: data ownership and communication overhead. This seminar will host four experts to provide insights on deploying a reliable and privacy-preserving edge intelligence solution from leading academic practitioners and industry partners. In particular, we will have two FL presentations, one presentation on the security of disconnecting IoT devices and one presentation on Graph federated data discovery and source.

📍 0:00 Introduction by the moderators, Ahmed Awad & Feras M. Awaysheh (UniTartuCS)
📍 15:38 Andreas Hellander (University of Uppsala): Scalable Federated Machine Learning with FEDn
📍 42:12 Peter Richtarik (KAUST): EF21: A new, simpler, theoretically better, and practically faster error feedback
📍 1:06:00 Aaron Ardiri (RIoT Secure): The Internet of Disconnected Things
📍 1:31:02 Essam Mansour (Concordia University): A Data Discovery Platform Empowered by Knowledge Graph Technologies: Challenges and Opportunities

Data Science Seminars are supported by the European Social Fund and University of Tartu ASTRA project PER ASPERA Doctoral School of Information and Communication Technologies.