The U.S. National Institute of Standards and Technology published guidance on privacy-preserving federated learning systems that are vertically partitioned. The guidance discusses different aspects of implementing vertical PPFL systems while taking steps to balance "the privacy risk of information leakage against the performance cost of preventing" data leaks.
NIST publishes guidance on privacy-preserving federated learning systems
Related stories
A view from Brussels: Will the EU pause the AI Act?
Notes from the Asia-Pacific region: India's digital landscape marches ahead
Navigate 2025: Rise of AI brings novel litigation, copyright issues
Navigate 2025: How individuals' feelings inform AI governance practices
There's no opting-out of universal opt-outs