Using AI-enabled Edge Cloud Technologies to Strengthen Europe’s...
summary
Shows a working MLOps stack that runs federated learning across factory edges to produce a shared global model without moving raw data, using Flower and MLflow and plugins for concept drift, model deployment and data signatures. Demonstrates two real pilots: camera X‑ray defect detection and time‑series models for rolling machines, both integrated into Mondragon assembly lines. Describes COGNIT’s distributed FaaS so IoT devices can offload tasks to the cloud‑edge continuum when needed.
AI-written summary · about
chapters (5)
- 0:00AI-enabled Edge Cloud intro
- 2:06Open source for AI deployment
- 3:54Federated learning for industry
- 5:59Sovereign Cloud Continuum for IoT
- 8:09Q&A on data trust
related talks
17mSponsored Keynote: AI-Driven Deployment and Monitoring Across Federated Cloud-Edge Sites
05mEdge AI: Because the Cloud is Too Slow...
14mEdge AI: Real Solutions for Real World Problems
19mDistributed AI for Automated and Scalable Quality Inspection using EDGE Computing...
33mAI on the Edge – Unlocking New Opportunities in IoT
15mUnlocking Networking AI Innovation: Introducing Essedum
15mSimplifying & Scaling Edge AI
38mHow AI Improves Operational Effectiveness, Safety and Well-Being