Beyond the Silo: Evaluating Multi-Party Machine Learning Technologies

Health privacy regulations, data residency rules, and trust concerns create data silos across the healthcare ecosystem that makes it hard to get the data you need to build powerful machine learning models. Multi-party machine learning technologies are opening up new opportunities for collaborations across sites, enabling data scientists to train new models with larger, more diverse data sets, while protecting patient privacy. But multi-party machine learning is a complex space with multiple technologies, from federated learning, to differential privacy, to homomorphic encryption, and beyond. How do you evaluate and decide which multi-party machine learning technologies are appropriate for your use case? I will be diving into the different technologies, and helping you ask the right questions to determine which technologies to start investing in.