The problem The previous post in our series discussed techniques for providing input privacy in PPFL systems where data is horizontally partitioned. This blog will focus on techniques for providing input privacy when data is vertically partitioned . As described in our third post , vertical partitioning is where the training data is divided across parties such that each party holds different columns of the data. In contrast to horizontally partitioned data, training a model on vertically partitioned data is more challenging as it is generally not possible to train separate models on different

The U.S. Small Business Administration is celebrating National Small Business Week from April 28 - May 4, 2024. This week recognizes and celebrates the small business community’s significant contributions to the nation. Organizations across the country participate by hosting in-person and virtual events, recognizing small business leaders and change-makers, and highlighting resources that help the small business community more easily and efficiently start and scale their businesses. To add to the festivities, this NIST Cybersecurity Insights blog showcases the NIST Cybersecurity Framework 2.0

We all need supplements sometimes. Whether it’s a little extra vitamin C during flu season or some vitamin D during the dark days of Winter. When used correctly, supplements help our body adjust to the changing conditions around us. Similarly, we are applying this same concept for the first time to our NIST SP 800-63B, Digital Identity Guidelines: Authentication and Lifecycle Management. Today, we published a supplement that provides interim guidance for agencies seeking to make use of ‘syncable authenticators’ ( for example, passkeys) in both enterprise-facing and public-facing use cases

In our second post we described attacks on models and the concepts of input privacy and output privacy. ln our last post, we described horizontal and vertical partitioning of data in privacy-preserving federated learning (PPFL) systems. In this post, we explore the problem of providing input privacy in PPFL systems for the horizontally-partitioned setting. Models, training, and aggregation To explore techniques for input privacy in PPFL, we first have to be more precise about the training process. In horizontally-partitioned federated learning, a common approach is to ask each participant to