OpenFL: A Federated Learning Framework for Secure Collaborative Model Training

Explore a comprehensive overview of OpenFL, a Python 3 framework for Federated Learning, in this informative conference talk. Discover how this flexible, extensible, and easily learnable tool enables organizations to collaboratively train models without sharing sensitive information. Learn about the project's community-driven approach, its narrow interfaces, and the ability to run processes within Trusted Execution Environments (TEE) for enhanced data and model confidentiality. Gain insights into a real-world application where Intel Labs and UPenn utilized data from over 71 medical institutions to test federated learning for brain tumor edge detection. Understand how federated learning hardware and software can secure sensitive data at the source while still benefiting from larger datasets. Find out how to adopt, contribute to, and secure federated learning projects using OpenFL.