U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

Privacy-Enhancing Technologies (PETs) Testbed

Organizations often cannot share sensitive data because of privacy, regulatory, and security requirements even though access to larger and more up-to-date datasets can improve research and analysis. The NCCoE PETs Testbed develops practical tools and guidelines that can help organizations evaluate technologies that protect sensitive data while enabling the data to be used for research and analysis. 

Giving researchers and organizations practical tools and guidelines to conduct structured, reproducible evaluations of Privacy-Enhancing Technologies.

By providing open-source testbeds, curated datasets, and reference architectures, the project enables organizations to experiment with PETs and evaluate the tradeoffs between privacy protection and data usefulness. Through this research, the NCCoE is accelerating the innovation and adoption of privacy-preserving technologies, while creating the foundations for future use of PETs to enable sensitive data analyses.
Status: Preparing Draft

To enable organizations to experiment with PETs, the project has released two resources:

  • Open-Source Testbed: An open-source software for training ML models centrally or across federated sites, with or without differential privacy. This tool allows users to compare how experiment configurations and privacy settings affect model performance. 
  • Curated Dataset: A machine learning-ready soybean genomic dataset covering over 18,000 accessions (or individual records) from the USDA Soybean Germplasm Collection. This serves as a common dataset to enable researchers to evaluate model performance across different testbed configurations.

For more details and information on how to use these resources, visit the PETs Testbeds page. To submit comments on the newly available resources, please email the team at [email protected].

Project Abstract

Machine learning (ML) models trained on sensitive and proprietary data can uncover new insights, for example in biomedical research, where patient data can help identify new treatments. However, organizations are often unable to share data because of privacy, regulatory, or security concerns.

PETs are technical controls that attempt to reduce privacy risks by limiting how data is collected, analyzed, and shared—which can enable secure collaboration.

Without a practical way to test or compare PETs, said organizations may struggle to determine wither certain PETs provide sufficient privacy protection while preserving analytic utility. 

The PETs Testbed provides tools and resources to investigate PETs and assess their suitability for specific use cases. The Testbed can help improve privacy risk management by giving stakeholders model environments, datasets, and metrics for examining both the benefits and limitations of PETs. 

Through this research, the NCCoE is accelerating the innovation and adoption of privacy-preserving technologies, while creating the foundations for future use of PETs to enable sensitive data analyses.