Data collaboration with Federated Learning
summary
Enterprise-grade federated learning platform that runs as one or two on-prem software nodes and stitches together homomorphic encryption and secure multi-party computation with PSI and TEEs to keep raw data local. Features include a data-market operator, request/approval workflow, traceable usage logs, column-name anonymization to protect IP, and Jupyter Notebook integration for easy adoption. Piloted with three major Hong Kong banks plus Science Park and TransUnion last year, delivering improved model accuracy for customer targeting and securities tasks while preserving compliance.
AI-written summary · about
chapters (11)
- 0:00Federated learning basics
- 1:08AI needs federated learning
- 3:14Data collaboration risks
- 8:37Secure collaboration with FL
- 10:03Privacy-enhancing tech overview
- 12:54Horizontal vs vertical FL
- 13:35FL industry use cases
- 15:42Enterprise FL solutions
- 21:50IP and privacy protection
- 23:13FL tools integration
- 24:19Q&A
related talks
42mFederated Learning: Unlocking the Values of Discrete Data for Enterprise
28mA Privacy Enhancing Federated Learning Framework for Multi-Party Distributed ..
31mFrom Theory to Practice: Advances in Real-World Federated Learning
22mCollaborative Self-Learning for Security, Fun and Profit
20mBeyond MLOps - How Federated MLOps Unlocks the Next Frontier of AI Innovation
29mBalancing Thin Line Between Data Intelligence & Privacy
21mWhat, Why and How of Federated Machine Learning – Implementation Using...
41mFederated Learning: Machine Learning on Decentralized Data