Federal Grant Award Summary

The University of North Carolina at Chapel Hill received a $175,000 Project Grant from the National Science Foundation's Division of Mathematical Sciences (Mathematical and Physical Sciences program, CFDA 47.049) beginning July 1, 2026, through June 30, 2029. This collaborative research initiative will develop the Binary Expansion Group Intersection Network (BEGIN), a novel statistical framework for learning from data at the binary digit level. The project will advance machine learning interpretability and statistical inference by constructing graphical models directly from binary representations of complex, multi-variable datasets, with applications across neuroscience, genetics, engineering, economics, and other scientific domains.

The research deliverables include establishing theoretical foundations and methodologies for testing and modeling conditional independence using bit-based approaches, leveraging concepts from binary expansion and Abelian group theory. Building on these foundations, the project will develop bit-based methods for causal inference and interpretable machine learning that are more reliable and broadly applicable than traditional statistical approaches. In addition to advancing the scientific foundations of data analysis and reproducibility in research, the project will provide comprehensive research training for graduate, undergraduate, and high school students.