Federated Learning from simulation to production
summary
NVFLARE is an Apache‑2 open source SDK that turns PyTorch/TensorFlow/Lightning training into federated jobs with 1–3 lines of code to patch trainers or add a decorator, runs a local simulator and a PC staging mode, and deploys to Azure/AWS. It ships built‑in HE‑style encryption and DP filters (SVT, Gaussian), enforces mTLS and per‑job signatures, encrypts models so servers cannot decrypt client data, supports federated XGBoost, GNNs and vertical FL, LLM prompt tuning with NeMo, and proves out on a COVID X‑ray study across 11 countries.
AI-written summary · about
chapters (7)
- 0:00Introduction to NVFlare
- 3:20Core NVFlare features
- 8:00Security and architecture
- 11:59Production readiness
- 17:41Customization and flexibility
- 21:14Adapting workflows to FL
- 24:32LLMs and live demonstration
related talks
31mFrom Theory to Practice: Advances in Real-World Federated Learning
27mFederated Learning with Azure Machine Learning NVIDIA FLARE and MONAI
13mScalable and Portable Federated Learning Simulation Engine
42mFederated Learning: Unlocking the Values of Discrete Data for Enterprise
31mFederated Learning: So Close and Yet So Out of Reach
26mOpenFL (Federated Learning) Building better AI models with private data
31mFedLess: Secure and Scalable Serverless Federated Learning
34mLFN AI Taskforce: Platform-Driven Fede...