Federal Grant Award Summary

Harvard College received a $175,000 Project Grant from the National Science Foundation (NSF) Division of Mathematical Sciences (CFDA 47.049) awarded on July 1, 2026, with a completion date of June 30, 2029. The award supports collaborative research on the Binary Expansion Group Intersection Network (BEGIN) framework, a novel statistical learning methodology that operates at the binary digit level of data representation. The project will develop theory and computational methods for constructing graphical models directly from binary data representations, utilizing concepts from binary expansion and Abelian group theory to analyze conditional dependence relationships among variables—the patterns of how variables interact after accounting for other variables' effects.

The research deliverables will establish foundational theory and methodology for testing and modeling conditional independence using bit-based representations, with subsequent applications to causal inference and interpretable machine learning. These advances are intended to produce more reliable, reproducible, and broadly applicable statistical tools for scientific discovery across multiple disciplines, including neuroscience, genetics, engineering, and economics. The project incorporates integrated research and educational components, providing research training opportunities for graduate students, undergraduate researchers, and high school participants at the Cambridge, Massachusetts campus location.