Federal Project Grant Summary

Massachusetts Institute of Technology received a $486,568 Project Grant from the National Science Foundation (NSF) Division of Physics under the Mathematical and Physical Sciences program (CFDA 47.049) effective October 1, 2026, through September 30, 2029. This collaborative research initiative delivers ML4GW, an open-source, PyTorch-based machine learning (ML) framework designed to accelerate gravitational wave (GW) data analysis. The framework provides graphics processing unit (GPU)-based tools that reduce alert latency to under one second, enable detection of astrophysical signals missed by traditional analysis methods, and decrease computational costs by orders of magnitude. The project directly addresses the computational demands of processing hundreds of gravitational wave events annually from modern detectors, with detection rates expected to increase to several events per day within the coming years.

The deliverables emphasize community accessibility and workforce development through open-source code, public datasets, pre-trained models, and educational resources targeting high school students through established researchers. By strengthening shared national computing infrastructure, the project benefits broader scientific disciplines including particle physics, neutrino astronomy, and time-domain astronomy. The award advances NSF's mission to promote mathematical and physical sciences research while enhancing infrastructure and broadening participation across the scientific workforce, particularly at smaller institutions and underrepresented groups in STEM fields.