Soledad Villar, assistant professor of applied mathematics and statistics, has been selected to receive the National Science Foundation (NSF)’s Early CAREER Award, which recognizes early stage scholars with high levels of promise and excellence.
Villar’s five-year project “Symmetries and Classical Physics in Machine Learning for Science and Engineering,” will blend of invariant theory, representation theory, differential geometry, and classical physics to facilitate advanced coordinate-free machine learning methods with a particular focus on applications in representation learning and physics emulation within the domains of cosmology and climate science.
Her research explores computational methods for extracting information from data with a focus on optimization for data science; machine learning and optimization; representation learning and graph neural networks; and equivariant machine learning.
She is a member of the Mathematical Institute for Data Science at Johns Hopkins.
This award excerpt was taken from the Mathematical Institute for Data Science.