FAR 13 Single Source or Brand Name Justification Less Than SAT Feasability Study - Redacted.pdf

United States Department of Agriculture Limited Source/Brand Name Justification

FAR 13.106-1(b)(1) – Simplified Acquisitions Control No. 1137438

1. Identification of the agency and the contracting activity.

Requesting Agency: Yang Cheng, Senior Mathematical Statistician Denise Abreu, Contracting Officer Representative

National Agricultural Statistics Service (NASS) 1400 Independence Avenue SW Washington, DC 20250

Contracting Activity: Katelyn Nelson, Contracting Officer 5601 Sunnyside Ave Beltsville, MD 20705

2. Description of supply or service required to meet the agency’s needs.

The United States Department of Agriculture (USDA), National Agricultural Statistics Service (NASS) is exploring innovative methodology to modernize the statistical disclosure limitation techniques applied to the Census of Agriculture and Survey data. Specifically, differential privacy (DP) is to be explored as a solution, and NASS wants to evaluate the feasibility of using DP.

NASS intends to award a sole source, firm fixed price contract, and as a result of the contract, NASS will receive a list of services from Tumult Labs.

• Provide progress reports and discuss questions raised by USDA NASS staff at weekly meetings.

• Develop a PowerPoint slide deck of a talk about the NASS disclosure feasibility study that can be presented at NASS and the broader statistical community; Tumult Labs and NASS will collaborate to present NASS disclosure feasibility studies at the 2024 FCSM research and policy conference.

• Create a slide deck or document to list findings and recommendations on USDA-NASS staff’s pseudo-code implementation.

• Train NASS staff in using Tumult Analytics.

• Assist USDA-NASS staff in their exploration of combining Per-Record differential privacy

(PRDP) with the cell suppression method.

• Evaluate at a high level whether the Random Tabular Adjustment-PRDP (RTA-PRDP) approach might have a formal mathematical grounding.

• Submit a summary report with some recommendations for future work on the continuous different privacy approach.

3. The rationale or circumstances for limiting competition.

Based on Tumult Labs’ work experience with the US Census Bureau regarding the Economic Census and the similarities between the Economic Census and the Census of Agriculture (COA), the NASS Statistical Disclosure Limitation (SDL) research team partnered with Tumult Labs to complete a feasibility study examining the application of per-record differential privacy to simulated Census of Agriculture Data. In this study, Tumult Labs defined an initial algorithm for using PRDP on the highly-skewed Census of Agriculture data, applied that algorithm on a set of simulated Census data that resembled 2017 Michigan non-citrus fruit and nut population estimates, and presented their findings to the SDL team and NASS management.

The work Tumult Labs has done with NASS and other government agencies in modernizing their disclosure avoidance methodologies to use differential privacy has demonstrated a highly unique capability that no other vendor has, and this capability is a perfect fit for USDA/NASS’s needs around evaluating differential privacy.

Tumult Labs has unique expertise in differential privacy, a promising disclosure limitation technique and is the market leader in providing differential privacy expertise and software tools for government agencies. The company was founded by three leading academics who made pivotal and award-winning research contributions to the field of differential privacy. Tumult founders and scientists were coauthors on a HDSR 2022 paper titled "The 2020 Census Disclosure Avoidance System Top Down Algorithm" that won the 2023 Caspar Bowden Award for Outstanding Research in Privacy Enhancing Technologies.

Tumult Labs’ founders were also the first to demonstrate how to deploy differential privacy in a government agency when they developed algorithms to publish U.S. Census Bureau OnTheMap data product with differential privacy in 2008.

Tumult Labs now has helped the National Agricultural Statistics Service (NASS) and several other US government agencies, including the Internal Revenue Service (IRS), evaluate and deploy differentially private statistical data products.

4. A description of the market research conducted and the results or a statement of the reason market research was not conducted.

During market research, it was discovered that three companies offer differential privacy as a service, including the following:

LeapYear 352 University AVE SW Unit S123 Atlanta, Georgia 30310

SAM UEI: L9GXPSQ9WSV4

Canopy Software 12020 Sunrise Valley Drive Suite #100 Reston, VA 20191

Privatar Victoria Station London SW1W 9RH United Kingdom

Neither Canopy nor Privitar are registered in SAM.gov, and of these three companies, the majority focus on industries in the field of healthcare, insurance, and data storage and generate the non-survey data from social media. Furthermore, none of these companies have experience applying differential privacy techniques to census or survey data in U.S. federal statistical agencies. Tumult Labs is the only company that has experience in applying differential privacy to not only census and survey data but has also provided solutions to federal statistical agencies. Therefore, Tumult Labs is the ONLY firm that can meet all requirements and special needs of NASS included in the Statement of Work (SOW).

5. Estimated price, including all options, if applicable.

The independent Government cost estimate is $38,000.00.

6. Recommended source(s).

Tumult Labs, Inc.

Ashwin Machanavajjhala ashwin@tmlt.io 201 W. Main Street Suite B26 Durham, NC 27701

SAM UEI: HW15BLJ3T9K1

Certifications

Supporting data that is the responsibility of technical or requirements personnel and which form the basis for this justification are accurate and complete to the best of my knowledge.

Yang Cheng, Ph.D.

Senior Mathematical Statistician USDA NASS Research and Development Division

The information contained in this justification is accurate and complete to the best of my knowledge and belief and in accordance with Federal Acquisition Regulation 13.106-1(b)(1), I hereby determine the circumstances described above deem only one source, or limited sources, reasonably available and competition is precluded for the reasons indicated in this document.

Katelyn Nelson Contracting Officer

The information contained in this justification is accurate and complete to the best of my knowledge and belief and in accordance with Federal Acquisition Regulation 13.106-1(b)(1), I hereby determine the circumstances described above deem only one source, or limited sources, reasonably available and competition is precluded for the reasons indicated in this document.

James Velez Contracting Officer Team Lead