Upcoming Events
CDS Colloquium -Synthetic Populations: From Privacy to Application
Oct 20, 2023, 3:00 - 4:00 PM
Center on Social Complexity, 3rd floor, Research Hall, Fairfax Campus and Virtual via Zoom
Speaker: Bianica Pires, Principal Modeling & Simulation Engineer, MITRE
Title: Synthetic Populations: From Privacy to Application
Abstract: Synthetic data are artificial data with the same general patterns and properties of the original data. The use of synthetic data is growing rapidly. Gartner predicts that by 2024, 60% of data used for AI and machine learning will be synthetic. Synthetic populations are a type of synthetic data designed to represent actual populations. They are a simplified representation because only attributes of interest are reproduced, and it is representative of the actual population because the statistical properties of the original population are maintained. Data providers such as the U.S. Census Bureau are particularly interested in synthetic data for preserving privacy due to the vulnerabilities associated with public use microdata in an evolving data landscape. From the perspective of the data analyst and modeler, synthetic populations provide a rich source of input into any analytics and modeling effort that would benefit from individual-level microdata. The first part of the talk will review current research with the Census to create an automated validation solution designed to maintain and build trust within the research community as Census moves in the direction of synthetic microdata. The second part of the talk will focus on the utility of synthetic populations for studying social systems. The Artificial Societies was developed to support both data analytics and dynamic modeling of synthetic populations. Examples across health, climate change, and urban land-use will be reviewed. © 2023 THE MITRE CORPORATION. ALL RIGHTS RESERVED. Approved for Public Release; Distribution Unlimited. Public Release Case Number 23-3419.
For Zoom link contact Robert Axtell