Algorithms for Threat Detection

The NSF's Algorithms for Threat Detection (ATD) program, in partnership with the National Geospatial Intelligence Agency (NGA), seeks to advance mathematical and statistical algorithms for analyzing large-scale spatiotemporal datasets. The program emphasizes detecting and mitigating threats through innovative approaches to data analysis, focusing on human dynamics and community activity patterns. Funded projects aim to identify anomalies and generate actionable insights from complex data, supporting national security and societal resilience goals. The program supports research in two primary areas: the development of novel computational algorithms for rapid and secure spatiotemporal data analysis, and mathematical frameworks guiding the application of advanced AI techniques, such as neural networks and large language models. Specific areas of interest include detecting hidden phenomena, improving AI reliability, quantifying uncertainty in data analysis, and developing methods for secure data integration and real-time processing. Eligibility extends to U.S.-based institutions of higher education, nonprofit organizations, and federally recognized tribal nations. Proposals must focus on research conducted in the U.S., although exceptions can be made for international campuses of U.S. institutions if justified. Principal Investigators (PIs) face no restrictions on proposal submission or participation. The anticipated funding is $3 million annually, with 10 to 20 awards expected each year. The program funds projects through standard or continuing grants, emphasizing high-risk, high-reward proposals and multidisciplinary collaborations. Cost-sharing is prohibited, and proposals should include budgets for participation in annual PI meetings. Proposals must comply with the NSF Proposal and Award Policies and Procedures Guide (PAPPG) and may be submitted via Research.gov or Grants.gov. The submission deadlines are April 10, 2024, February 19, 2025, and the third Wednesday of February annually thereafter. Proposals are evaluated based on intellectual merit, broader societal impacts, and alignment with program goals. Awardees are required to submit regular progress reports through Research.gov, maintain compliance with NSF policies, and acknowledge support from the NSF and NGA in publications. Additional reporting requirements include a final project report summarizing research outcomes and societal impacts. The program strongly encourages diverse participation and innovative research methods to address complex national security challenges through advanced analytics and computational science.