Scalable and Portable Federated Learning Simulation Engine
summary
A scalable, portable, hardware‑independent simulation engine called FLSE runs federated learning experiments end to end, packing Flower framework integrations into a multicontainer executor cluster with a broker that dispatches tasks to parallel executors. It automates training, verification and validation across ad hoc scenarios, reports per‑scenario metrics, and flags resource or model constraints; used on European healthcare servers, it proves feasibility but requires model restrictions (not every model is auto‑deployable).
AI-written summary · about
chapters (9)
- 0:00Introduction
- 0:43Federated learning basics
- 2:50Why federated learning
- 3:53Federated learning challenges
- 5:16Simulation tool introduction
- 7:04Simulation benefits
- 8:33Technical architecture
- 9:37Model integration support
- 12:46Q&A
related talks
32mInnovation in the Age of Regulation: Federated Learning with Flower
30mFederated Learning from simulation to production
57mKeynote: Federated Learning with Flower: AI's Next Frontier
42mFederated Learning: Unlocking the Values of Discrete Data for Enterprise
01h 22mThe future of AI training is federated
32mOpenFL: A Federated Learning Project to Power (and Secure) Your Projects
32mPlant a Touch-Me-Not: Train Models Without Anyone Touching Your Data with Flower
31mFedLess: Secure and Scalable Serverless Federated Learning