Project Grant 2337678

This $120,000 federal Project Grant award from the National Science Foundation's Mathematical and Physical Sciences (CFDA 47.049) program supports research at the University of Washington to develop mathematical frameworks, algorithms, and computational methodologies for scientific machine learning using Gaussian processes. The key focus areas are: (1) using Gaussian processes to solve nonlinear, high-dimensional, and parametric partial differential equations; (2) Gaussian process-based methods for operator learning, emulation, and physics discovery; and (3) Gaussian processes for high-dimensional sampling, inference, and generative modeling. The 5-year project aims to advance the theoretical understanding and practical capabilities of machine learning approaches in scientific computing applications, while also incorporating an extensive education plan for training students from high school through graduate levels. No sub-awards are planned under this grant, which was awarded on June 1, 2024 with an expected completion date of May 31, 2029.